diff --git a/NAMESPACE b/NAMESPACE
index 323e34ba..4f9a1766 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -1,9 +1,11 @@
 # Generated by roxygen2: do not edit by hand
 
 S3method(as_continuous_pal,"function")
+S3method(as_continuous_pal,character)
 S3method(as_continuous_pal,default)
 S3method(as_continuous_pal,pal_discrete)
 S3method(as_discrete_pal,"function")
+S3method(as_discrete_pal,character)
 S3method(as_discrete_pal,default)
 S3method(as_discrete_pal,pal_continuous)
 S3method(fullseq,Date)
@@ -50,6 +52,7 @@ export(asinh_trans)
 export(asn_trans)
 export(atanh_trans)
 export(boxcox_trans)
+export(breaks_exp)
 export(breaks_extended)
 export(breaks_log)
 export(breaks_pretty)
@@ -96,7 +99,9 @@ export(exp_trans)
 export(expand_range)
 export(extended_breaks)
 export(format_format)
+export(format_log)
 export(fullseq)
+export(get_palette)
 export(gradient_n_pal)
 export(grey_pal)
 export(hms_trans)
@@ -138,6 +143,7 @@ export(log_trans)
 export(logit_trans)
 export(manual_pal)
 export(math_format)
+export(minor_breaks_log)
 export(minor_breaks_n)
 export(minor_breaks_width)
 export(modulus_trans)
@@ -149,6 +155,7 @@ export(number)
 export(number_bytes)
 export(number_bytes_format)
 export(number_format)
+export(number_options)
 export(oob_censor)
 export(oob_censor_any)
 export(oob_discard)
@@ -176,6 +183,7 @@ export(pal_seq_gradient)
 export(pal_shape)
 export(pal_viridis)
 export(palette_na_safe)
+export(palette_names)
 export(palette_nlevels)
 export(palette_type)
 export(parse_format)
@@ -194,10 +202,12 @@ export(rescale_max)
 export(rescale_mid)
 export(rescale_none)
 export(rescale_pal)
+export(reset_palettes)
 export(reverse_trans)
 export(scientific)
 export(scientific_format)
 export(seq_gradient_pal)
+export(set_palette)
 export(shape_pal)
 export(show_col)
 export(sqrt_trans)
diff --git a/NEWS.md b/NEWS.md
index 9e2f72f3..21c97925 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -1,6 +1,13 @@
 # scales (development version)
 
 * New `label_date_short(leading)` argument to replace leading zeroes (#442)
+* `breaks_pretty()` will return the input limit when it has no range (#446)
+* `transform_exp()` now has more sensible breaks, available in `breaks_exp()` 
+  (@teunbrand, #405).
+* The scales package now keeps track of known palettes. These can be retrieved
+  using `get_palette()` or registered using `set_palette()` (#396).
+* `label_log()` has a `signed` argument for displaying negative numbers 
+  (@teunbrand, #421).
 
 # scales 1.3.0
 
diff --git a/R/breaks-log.R b/R/breaks-log.R
index 79acceb4..cbfe7cfc 100644
--- a/R/breaks-log.R
+++ b/R/breaks-log.R
@@ -82,6 +82,89 @@ breaks_log <- function(n = 5, base = 10) {
 #' @rdname breaks_log
 log_breaks <- breaks_log
 
+#' Minor breaks for log-10 axes
+#'
+#' This break function is designed to mark every power, multiples of 5 and/or 1
+#' of that power for base 10.
+#'
+#' @param detail Any of `1`, `5` and `10` to mark multiples of
+#'   powers, multiples of 5 of powers or just powers respectively.
+#' @param smallest Smallest absolute value to mark when the range includes
+#'   negative numbers.
+#'
+#' @return A function to generate minor ticks.
+#' @export
+#'
+#' @examples
+#' # Standard usage with log10 scale
+#' demo_log10(c(1, 1e10), minor_breaks = minor_breaks_log())
+#' # Increasing detail over many powers
+#' demo_log10(c(1, 1e10), minor_breaks = minor_breaks_log(detail = 1))
+#' # Adjusting until where to draw minor breaks
+#' demo_continuous(
+#'   c(-1000, 1000),
+#'   transform = asinh_trans(),
+#'   minor_breaks = minor_breaks_log(smallest = 1)
+#' )
+minor_breaks_log <- function(detail = NULL, smallest = NULL) {
+  if (!is.null(detail) && (!length(detail) == 1 || !detail %in% c(1, 5, 10))) {
+    cli::cli_abort("The {.arg detail} argument must be one of 1, 5 or 10.")
+  }
+  if (!is.null(smallest) &&
+      (!length(smallest) == 1 || smallest < 1e-100 || !is.finite(smallest))) {
+    cli::cli_abort(
+      "The {.arg smallest} argument must be a finite, positive, non-zero number."
+    )
+  }
+  force(smallest)
+  function(x, ...) {
+
+    has_negatives <- any(x <= 0)
+
+    if (has_negatives) {
+      large <- max(abs(x))
+      small <- smallest %||% min(c(1, large) * 0.1)
+      x <- sort(c(small * 10, large))
+    }
+
+    start <- floor(log10(min(x))) - 1L
+    end   <- ceiling(log10(max(x))) + 1L
+
+    if (is.null(detail)) {
+      i <- findInterval(abs(end - start), c(8, 15), left.open = TRUE) + 1L
+      detail <- c(1, 5, 10)[i]
+    }
+
+    ladder <- 10^seq(start, end, by = 1L)
+    tens <- fives <- ones <- numeric()
+    if (detail %in% c(10, 5, 1)) {
+      tens <- ladder
+    }
+    if (detail %in% c(5, 1)) {
+      fives <- 5 * ladder
+    }
+    if (detail == 1) {
+      ones <- as.vector(outer(1:9, ladder))
+      ones <- setdiff(ones, c(tens, fives))
+    }
+
+    if (has_negatives) {
+      tens  <- tens[tens >= small]
+      tens  <- c(tens, -tens, 0)
+      fives <- fives[fives >= small]
+      fives <- c(fives, -fives)
+      ones  <- ones[ones >= small]
+      ones  <- c(ones, -ones)
+    }
+
+    ticks <- c(tens, fives, ones)
+    n <- c(length(tens), length(fives), length(ones))
+
+    attr(ticks, "detail") <- rep(c(10, 5, 1), n)
+    ticks
+  }
+}
+
 #' @author Thierry Onkelinx, \email{thierry.onkelinx@inbo.be}
 #' @noRd
 log_sub_breaks <- function(rng, n = 5, base = 10) {
diff --git a/R/breaks.R b/R/breaks.R
index 947f3aa6..adf194d2 100644
--- a/R/breaks.R
+++ b/R/breaks.R
@@ -122,6 +122,9 @@ breaks_pretty <- function(n = 5, ...) {
   force_all(n, ...)
   n_default <- n
   function(x, n = n_default) {
+    if (zero_range(as.numeric(x))) {
+      return(x[1])
+    }
     breaks <- pretty(x, n, ...)
     names(breaks) <- attr(breaks, "labels")
     breaks
@@ -182,3 +185,31 @@ breaks_timespan <- function(unit = c("secs", "mins", "hours", "days", "weeks"),
     as.difftime(breaks * scale, units = "secs")
   }
 }
+
+#' Breaks for exponentially transformed data
+#'
+#' This breaks function typically labels zero and the last `n - 1` integers of a
+#' range if that range is large enough (currently: 3). For smaller ranges, it
+#' uses [`breaks_extended()`].
+#'
+#' @inheritParams breaks_extended
+#' @export
+#' @examples
+#' # Small range
+#' demo_continuous(c(100, 102), transform = "exp", breaks = breaks_exp())
+#' # Large range
+#' demo_continuous(c(0, 100), transform = "exp", breaks = breaks_exp(n = 4))
+breaks_exp <- function(n = 5, ...) {
+  n_default <- n
+  default <- extended_breaks(n = n_default, ...)
+  function(x, n = n_default) {
+    # Discard -Infs
+    x <- sort(pmax(x, 0))
+    top <- floor(x[2])
+    if (top >= 3 && abs(diff(x)) >= 3) {
+      unique(c(top - seq_len(min(top, n_default - 1)) + 1, 0))
+    } else {
+      default(x)
+    }
+  }
+}
diff --git a/R/colour-manip.R b/R/colour-manip.R
index 5de30897..c4f15d85 100644
--- a/R/colour-manip.R
+++ b/R/colour-manip.R
@@ -87,6 +87,14 @@ alpha <- function(colour, alpha = NA) {
 show_col <- function(colours, labels = TRUE, borders = NULL, cex_label = 1,
                      ncol = NULL) {
   n <- length(colours)
+  if (n == 1 && (is.function(colours) || !is_color(colours))) {
+    colours <- as_discrete_pal(colours)
+    n <- palette_nlevels(colours)
+    n <- if (is.na(n)) 16 else n
+    colours <- colours(n = n)
+    n <- length(colours)
+  }
+
   ncol <- ncol %||% ceiling(sqrt(length(colours)))
   nrow <- ceiling(n / ncol)
 
diff --git a/R/label-currency.R b/R/label-currency.R
index 56979a2f..98d18e3d 100644
--- a/R/label-currency.R
+++ b/R/label-currency.R
@@ -48,8 +48,11 @@
 #'   scale_cut = c(0, k = 1e3, m = 1e6, bn = 1e9, tn = 1e12)
 #' )
 #' demo_log10(c(1, 1e12), breaks = log_breaks(5, 1e3), labels = gbp)
-label_currency <- function(accuracy = NULL, scale = 1, prefix = "$",
-                           suffix = "", big.mark = ",", decimal.mark = ".",
+label_currency <- function(accuracy = NULL, scale = 1,
+                           prefix = NULL,
+                           suffix = NULL,
+                           big.mark = NULL,
+                           decimal.mark = NULL,
                            trim = TRUE, largest_with_fractional = 100000,
                            ...) {
   force_all(
@@ -144,13 +147,17 @@ dollar_format <- function(accuracy = NULL, scale = 1, prefix = "$",
 #' @export
 #' @rdname dollar_format
 #' @param x A numeric vector
-dollar <- function(x, accuracy = NULL, scale = 1, prefix = "$",
-                   suffix = "", big.mark = ",", decimal.mark = ".",
+dollar <- function(x, accuracy = NULL, scale = 1, prefix = NULL,
+                   suffix = NULL, big.mark = NULL, decimal.mark = NULL,
                    trim = TRUE, largest_with_cents = 100000,
                    negative_parens = deprecated(),
                    style_negative = c("hyphen", "minus", "parens"),
                    scale_cut = NULL,
                    ...) {
+  prefix <- prefix %||% getOption("scales.currency.prefix", default = "$")
+  suffix <- suffix %||% getOption("scales.currency.suffix", default = "")
+  big.mark <- big.mark %||% getOption("scales.currency.big.mark", default = ",")
+  decimal.mark <- decimal.mark %||% getOption("scales.currency.decimal.mark", default = ".")
   if (length(x) == 0) {
     return(character())
   }
diff --git a/R/label-date.R b/R/label-date.R
index adb82262..b2735eac 100644
--- a/R/label-date.R
+++ b/R/label-date.R
@@ -11,10 +11,10 @@
 #' suffix to the input (ns, us, ms, s, m, h, d, w).
 #'
 #' @inherit label_number return
-#' @param format For `date_format()` and `time_format()` a date/time format
+#' @param format For `label_date()` and `label_time()` a date/time format
 #'   string using standard POSIX specification.  See [strptime()] for details.
 #'
-#'   For `date_short()` a character vector of length 4 giving the format
+#'   For `label_date_short()` a character vector of length 4 giving the format
 #'   components to use for year, month, day, and hour respectively.
 #' @param tz a time zone name, see [timezones()]. Defaults
 #'   to UTC
@@ -33,9 +33,9 @@
 #'
 #' two_months <- date_range("2020-05-01", 60)
 #' demo_datetime(two_months)
-#' demo_datetime(two_months, labels = date_format("%m/%d"))
-#' demo_datetime(two_months, labels = date_format("%e %b", locale = "fr"))
-#' demo_datetime(two_months, labels = date_format("%e %B", locale = "es"))
+#' demo_datetime(two_months, labels = label_date("%m/%d"))
+#' demo_datetime(two_months, labels = label_date("%e %b", locale = "fr"))
+#' demo_datetime(two_months, labels = label_date("%e %B", locale = "es"))
 #' # ggplot2 provides a short-hand:
 #' demo_datetime(two_months, date_labels = "%m/%d")
 #'
diff --git a/R/label-log.R b/R/label-log.R
index 802d2a13..e31e2bb8 100644
--- a/R/label-log.R
+++ b/R/label-log.R
@@ -1,10 +1,18 @@
 #' Label numbers in log format (10^3, 10^6, etc)
 #'
-#' `label_log()` displays numbers as base^exponent, using superscript formatting.
+#' `label_log()` and `format_log()` display numbers as base^exponent, using
+#' superscript formatting. `label_log()` returns expressions suitable for
+#' labelling in scales, whereas `format_log()` returns deparsed text.
 #'
+#'
+#' @param x A numeric vector to format
 #' @param base Base of logarithm to use
 #' @param digits Number of significant digits to show for the exponent. Argument
 #'   is passed on to [base::format()].
+#' @param signed Should a `+` or `-` be displayed as a prefix? The
+#'   default, `NULL`, displays signs if there are zeroes or negative numbers
+#'   present.
+#' @param ... Passed on to `format()`.
 #' @inherit label_number return
 #' @seealso [breaks_log()] for the related breaks algorithm.
 #' @export
@@ -12,19 +20,43 @@
 #' @examples
 #' demo_log10(c(1, 1e5), labels = label_log())
 #' demo_log10(c(1, 1e5), breaks = breaks_log(base = 2), labels = label_log(base = 2))
-label_log <- function(base = 10, digits = 3) {
+#' format_log(c(0.1, 1, 10))
+label_log <- function(base = 10, digits = 3, signed = NULL) {
   function(x) {
-    if (length(x) == 0) {
-      return(expression())
-    }
-
-    exponent <- format(log(x, base = base), digits = digits)
-    text <- paste0(base, "^", exponent)
+    text <- format_log(x, base = base, signed = signed, digits = digits)
     ret <- parse_safe(text)
-
     # restore NAs from input vector
     ret[is.na(x)] <- NA
-
     ret
   }
 }
+
+#' @export
+#' @rdname label_log
+format_log <- function(x, base = 10, signed = NULL, ...) {
+
+  if (length(x) == 0) {
+    return(character())
+  }
+  prefix <- rep("", length(x))
+  finites <- x[is.finite(x)]
+
+  signed <- signed %||% any(finites <= 0)
+  if (signed) {
+    sign <- sign(x)
+    prefix[sign == +1] <- "+"
+    prefix[sign == -1] <- "-"
+    x <- abs(x)
+    x[x == 0] <- 1
+  }
+
+  exponent <- format(zapsmall(log(x, base = base)), ...)
+  text <- paste0(prefix, base, "^", exponent)
+
+  if (signed) {
+    text[sign == 0] <- "0"
+  }
+  text[is.na(x)] <- NA
+
+  text
+}
diff --git a/R/label-number.R b/R/label-number.R
index a8a335ae..b5dd04de 100644
--- a/R/label-number.R
+++ b/R/label-number.R
@@ -28,8 +28,11 @@
 #'   processed so that `prefix = "$"` will yield (e.g.) `-$1` and `($1)`.
 #' @param suffix Additional text to display after the number.
 #' @param big.mark Character used between every 3 digits to separate thousands.
+#'   The default (`NULL`) retrieves the setting from the
+#'   [number options][number_options].
 #' @param decimal.mark The character to be used to indicate the numeric
-#'   decimal point.
+#'   decimal point.  The default (`NULL`) retrieves the setting from the
+#'   [number options][number_options].
 #' @param style_positive A string that determines the style of positive numbers:
 #'
 #'   * `"none"` (the default): no change, e.g. `1`.
@@ -38,6 +41,9 @@
 #'     as wide as a number or `+`. Compared to `"none"`, adding a figure space
 #'     can ensure numbers remain properly aligned when they are left- or
 #'     right-justified.
+#'
+#'   The default (`NULL`) retrieves the setting from the
+#'   [number options][number_options].
 #' @param style_negative A string that determines the style of negative numbers:
 #'
 #'   * `"hyphen"` (the default): preceded by a standard hypen `-`, e.g. `-1`.
@@ -45,6 +51,9 @@
 #'      nicety that ensures `-` aligns with the horizontal bar of the
 #'      the horizontal bar of `+`.
 #'   * `"parens"`, wrapped in parentheses, e.g. `(1)`.
+#'
+#'   The default (`NULL`) retrieves the setting from the
+#'   [number options][number_options].
 #' @param scale_cut Named numeric vector that allows you to rescale large
 #'   (or small) numbers and add a prefix. Built-in helpers include:
 #'   * `cut_short_scale()`: [10^3, 10^6) = K, [10^6, 10^9) = M, [10^9, 10^12) = B, [10^12, Inf) = T.
@@ -109,9 +118,9 @@
 #' demo_continuous(c(32, 212), labels = label_number(suffix = "\u00b0F"))
 #' demo_continuous(c(0, 100), labels = label_number(suffix = "\u00b0C"))
 label_number <- function(accuracy = NULL, scale = 1, prefix = "",
-                         suffix = "", big.mark = " ", decimal.mark = ".",
-                         style_positive = c("none", "plus", "space"),
-                         style_negative = c("hyphen", "minus", "parens"),
+                         suffix = "", big.mark = NULL, decimal.mark = NULL,
+                         style_positive = NULL,
+                         style_negative = NULL,
                          scale_cut = NULL,
                          trim = TRUE, ...) {
   force_all(
@@ -225,17 +234,21 @@ comma_format <- label_comma
 #' @inheritParams label_number
 #' @return A character vector of `length(x)`.
 number <- function(x, accuracy = NULL, scale = 1, prefix = "",
-                   suffix = "", big.mark = " ", decimal.mark = ".",
-                   style_positive = c("none", "plus", "space"),
-                   style_negative = c("hyphen", "minus", "parens"),
+                   suffix = "", big.mark = NULL, decimal.mark = NULL,
+                   style_positive = NULL,
+                   style_negative = NULL,
                    scale_cut = NULL,
                    trim = TRUE, ...) {
   if (length(x) == 0) {
     return(character())
   }
+  big.mark <- big.mark %||% getOption("scales.big.mark", default = " ")
+  decimal.mark <- decimal.mark %||% getOption("scales.decimal.mark", default = ".")
+  style_positive <- style_positive %||% getOption("scales.style_positive", default = "none")
+  style_negative <- style_negative %||% getOption("scales.style_negative", default = "hyphen")
 
-  style_positive <- arg_match(style_positive)
-  style_negative <- arg_match(style_negative)
+  style_positive <- arg_match(style_positive, c("none", "plus", "space"))
+  style_negative <- arg_match(style_negative, c("hyphen", "minus", "parens"))
 
   if (!is.null(scale_cut)) {
     cut <- scale_cut(x,
@@ -298,6 +311,58 @@ number <- function(x, accuracy = NULL, scale = 1, prefix = "",
   ret
 }
 
+#' Number options
+#'
+#' Control the settings for formatting numbers globally.
+#'
+#' @inheritParams label_number
+#' @param currency.prefix,currency.suffix,currency.decimal.mark,currency.big.mark
+#'   Settings for [`label_currency()`] passed on without the `currency.`-prefix.
+#' @param ordinal.rules Setting for [`label_ordinal()`] passed on without the
+#'   `ordinal.`-prefix.
+#'
+#' @return The old options invisibly
+#' @export
+#'
+#' @examples
+#' # Default number formatting
+#' x <- c(0.1, 1, 1000)
+#' label_number()(x)
+#'
+#' # Now again with new options set
+#' number_options(style_positive = "plus", decimal.mark = ",")
+#' label_number()(x)
+#'
+#' # The options are the argument names with a 'scales.'-prefix
+#' options("scales.style_positive")
+#'
+#' # Resetting the options to their defaults
+#' number_options()
+#' label_number()(x)
+number_options <- function(
+  decimal.mark = ".",
+  big.mark = " ",
+  style_positive = c("none", "plus", "space"),
+  style_negative = c("hyphen", "minus", "parens"),
+  currency.prefix = "$",
+  currency.suffix = "",
+  currency.decimal.mark = decimal.mark,
+  currency.big.mark = setdiff(c(".", ","), currency.decimal.mark)[1],
+  ordinal.rules = ordinal_english()
+) {
+  opts <- options(
+    scales.decimal.mark          = decimal.mark,
+    scales.big.mark              = big.mark,
+    scales.style_positive        = arg_match(style_positive),
+    scales.style_negative        = arg_match(style_negative),
+    scales.currency.prefix       = currency.prefix,
+    scales.currency.suffix       = currency.suffix,
+    scales.currency.decimal.mark = currency.decimal.mark,
+    scales.currency.big.mark     = currency.big.mark,
+    scales.ordinal.rules         = ordinal.rules
+  )
+  invisible(opts)
+}
 
 # Helpers -----------------------------------------------------------------
 
@@ -350,6 +415,7 @@ scale_cut <- function(x, breaks, scale = 1, accuracy = NULL, suffix = "") {
   if (any(bad_break)) {
     # If the break below result in a perfect cut, prefer it
     lower_break <- breaks[match(break_suffix[bad_break], names(breaks)) - 1]
+    lower_break[lower_break == 0] <- 1  # Avoid choosing a non-existent break
     improved_break <- (x[bad_break] * scale / lower_break) %% 1 == 0
     # Unless the break below is a power of 10 change (1.25 is as good as 1250)
     power10_break <- log10(breaks[break_suffix[bad_break]] / lower_break) %% 1 == 0
diff --git a/R/label-ordinal.R b/R/label-ordinal.R
index b9de36f4..c4cdeafc 100644
--- a/R/label-ordinal.R
+++ b/R/label-ordinal.R
@@ -28,8 +28,8 @@
 #'   labels = label_ordinal(),
 #'   breaks = breaks_width(2)
 #' )
-label_ordinal <- function(prefix = "", suffix = "", big.mark = " ",
-                          rules = ordinal_english(), ...) {
+label_ordinal <- function(prefix = "", suffix = "", big.mark = NULL,
+                          rules = NULL, ...) {
   force_all(prefix, suffix, big.mark, rules, ...)
   function(x) {
     ordinal(
@@ -96,7 +96,9 @@ ordinal_format <- label_ordinal
 #' @export
 #' @rdname ordinal_format
 ordinal <- function(x, prefix = "", suffix = "", big.mark = " ",
-                    rules = ordinal_english(), ...) {
+                    rules = NULL, ...) {
+
+  rules <- rules %||% getOption("scales.ordinal.rules", default = ordinal_english())
   na_idx <- is.na(x)
   x <- round(x, digits = 0)
   x[na_idx] <- 1 # replace NAs with dummy value
diff --git a/R/label-percent.R b/R/label-percent.R
index 5e482d31..2dbd3d7d 100644
--- a/R/label-percent.R
+++ b/R/label-percent.R
@@ -15,7 +15,7 @@
 #' )
 #' demo_continuous(c(0, .01), labels = french_percent)
 label_percent <- function(accuracy = NULL, scale = 100, prefix = "",
-                          suffix = "%", big.mark = " ", decimal.mark = ".",
+                          suffix = "%", big.mark = NULL, decimal.mark = NULL,
                           trim = TRUE, ...) {
   number_format(
     accuracy = accuracy,
@@ -45,7 +45,7 @@ percent_format <- label_percent
 #' @export
 #' @rdname percent_format
 percent <- function(x, accuracy = NULL, scale = 100, prefix = "",
-                    suffix = "%", big.mark = " ", decimal.mark = ".",
+                    suffix = "%", big.mark = NULL, decimal.mark = NULL,
                     trim = TRUE, ...) {
   number(
     x = x,
diff --git a/R/label-pvalue.R b/R/label-pvalue.R
index 2564daf0..2b6a328b 100644
--- a/R/label-pvalue.R
+++ b/R/label-pvalue.R
@@ -4,8 +4,8 @@
 #'
 #' @inherit label_number return params
 #' @param prefix A character vector of length 3 giving the prefixes to
-#'   put in front of numbers. The default values are `c("<", "", ">")`
-#'   if `add_p` is `TRUE` and `c("p<", "p=", "p>")` if `FALSE`.
+#'   put in front of numbers. The default values are `c("p<", "p=", "p>")`
+#'   if `add_p` is `TRUE` and `c("<", "", ">")` if `FALSE`.
 #' @param add_p Add "p=" before the value?
 #' @export
 #' @family labels for continuous scales
@@ -18,7 +18,7 @@
 #' # Or provide your own prefixes
 #' prefix <- c("p < ", "p = ", "p > ")
 #' demo_continuous(c(0, 1), labels = label_pvalue(prefix = prefix))
-label_pvalue <- function(accuracy = .001, decimal.mark = ".", prefix = NULL, add_p = FALSE) {
+label_pvalue <- function(accuracy = .001, decimal.mark = NULL, prefix = NULL, add_p = FALSE) {
   force_all(accuracy, decimal.mark, add_p)
   function(x) {
     pvalue(
@@ -49,9 +49,10 @@ pvalue_format <- label_pvalue
 #' @export
 pvalue <- function(x,
                    accuracy = .001,
-                   decimal.mark = ".",
+                   decimal.mark = NULL,
                    prefix = NULL,
                    add_p = FALSE) {
+  decimal.mark <- decimal.mark %||% getOption("scales.decimal.mark", default = ".")
   out <- number(x, accuracy, decimal.mark = decimal.mark)
   below <- number(accuracy, accuracy, decimal.mark = decimal.mark)
   above <- number(1 - accuracy, accuracy, decimal.mark = decimal.mark)
diff --git a/R/label-scientific.R b/R/label-scientific.R
index 694073d9..ade16cad 100644
--- a/R/label-scientific.R
+++ b/R/label-scientific.R
@@ -14,7 +14,7 @@
 #'
 #' demo_log10(c(1, 1e9))
 label_scientific <- function(digits = 3, scale = 1, prefix = "", suffix = "",
-                             decimal.mark = ".", trim = TRUE, ...) {
+                             decimal.mark = NULL, trim = TRUE, ...) {
   force_all(digits, scale, prefix, suffix, decimal.mark, trim, ...)
   function(x) {
     scientific(
@@ -45,11 +45,12 @@ scientific_format <- label_scientific
 #' @export
 #' @rdname scientific_format
 scientific <- function(x, digits = 3, scale = 1, prefix = "", suffix = "",
-                       decimal.mark = ".", trim = TRUE, ...) {
+                       decimal.mark = NULL, trim = TRUE, ...) {
   if (length(x) == 0) {
     return(character())
   }
   x <- signif(x * scale, digits)
+  decimal.mark <- decimal.mark %||% getOption("scales.decimal.mark", default = ".")
   ret <- paste0(
     prefix,
     format(x, decimal.mark = decimal.mark, trim = trim, scientific = TRUE, ...),
diff --git a/R/labels-retired.R b/R/labels-retired.R
index be1e2fe6..62026af1 100644
--- a/R/labels-retired.R
+++ b/R/labels-retired.R
@@ -111,7 +111,7 @@ trans_format <- function(trans, format = scientific_format()) {
 #' demo_continuous(c(0, 2500), labels = km)
 unit_format <- function(accuracy = NULL, scale = 1, prefix = "",
                         unit = "m", sep = " ", suffix = paste0(sep, unit),
-                        big.mark = " ", decimal.mark = ".",
+                        big.mark = NULL, decimal.mark = NULL,
                         trim = TRUE, ...) {
   number_format(
     accuracy = accuracy,
diff --git a/R/pal-.R b/R/pal-.R
index 46e34573..8665dfc6 100644
--- a/R/pal-.R
+++ b/R/pal-.R
@@ -156,6 +156,14 @@ as_discrete_pal.pal_continuous <- function(x, ...) {
   )
 }
 
+#' @export
+as_discrete_pal.character <- function(x, ...) {
+  if (length(x) > 1)  {
+    return(pal_manual(x))
+  }
+  as_discrete_pal(get_palette(x, ...))
+}
+
 ## As continuous palette --------------------------------------------------
 
 #' @rdname new_continuous_palette
@@ -197,3 +205,11 @@ as_continuous_pal.pal_discrete <- function(x, ...) {
     )
   )
 }
+
+#' @export
+as_continuous_pal.character <- function(x, ...) {
+  if (length(x) > 1) {
+    return(colour_ramp(x))
+  }
+  as_continuous_pal(get_palette(x, ...))
+}
diff --git a/R/pal-shape.r b/R/pal-shape.r
index 7090631e..065e0c0f 100644
--- a/R/pal-shape.r
+++ b/R/pal-shape.r
@@ -8,7 +8,7 @@ pal_shape <- function(solid = TRUE) {
     if (n > 6) {
       cli::cli_warn(c(
         "The shape palette can deal with a maximum of 6 discrete values because more than 6 becomes difficult to discriminate",
-        i = "you have requested {n} values. Consider specifying shapes manually if you need that many have them."
+        i = "you have requested {n} values. Consider specifying shapes manually if you need that many of them."
       ))
     }
 
diff --git a/R/palette-registry.R b/R/palette-registry.R
new file mode 100644
index 00000000..6ef38ab7
--- /dev/null
+++ b/R/palette-registry.R
@@ -0,0 +1,157 @@
+.known_palettes <- new_environment(parent = empty_env())
+
+#' Known palettes
+#'
+#' The scales packages keeps track of a set of palettes it considers 'known'.
+#' The benefit of a known palette is that it can be called by name in functions
+#' as `as_continuous_pal()` or `as_discrete_pal()`.
+#'
+#' @param name A string giving the palette name.
+#' @param palette A [palette][new_continuous_palette], `function` or character
+#'   vector.
+#' @param warn_conflict A boolean which if `TRUE` (default), warns when
+#'   replacing a known palette.
+#' @param ... Additional arguments to pass to palette when it is a function
+#'   but not a palette class function.
+#'
+#' @return The `get_palette()` function returns a palette. The `set_palette()`
+#'   function is called for side effects and returns nothing.
+#' @export
+#'
+#' @examples
+#' # Get one of the known palettes
+#' get_palette("hue")
+#'
+#' # Set a new custom palette
+#' cols <- c("palegreen", "deepskyblue", "magenta")
+#' set_palette("aurora", palette = cols)
+#'
+#' # Palette is now known
+#' "aurora" %in% palette_names()
+#' as_continuous_pal("aurora")
+#'
+#' # Resetting palettes
+#' reset_palettes()
+get_palette <- function(name, ...) {
+
+  name <- tolower(name)
+  if (!exists(name, envir = .known_palettes)) {
+    cli::cli_abort("Unknown palette: {name}")
+  }
+
+  pal <- env_get(.known_palettes, name)
+
+  # Palette could be factory, in which case we want the product, or
+  # palette can be a palette function that isn't registered as such,
+  # in which case we want the colours it gives
+  if (is_function(pal) && !is_pal(pal)) {
+    pal <- try_fetch(
+      pal(...),
+      error = function(cnd) {
+        cli::cli_abort("Failed to interpret {name} as palette.", parent = cnd)
+      }
+    )
+  }
+  if (is.character(pal)) {
+    pal <- manual_pal(pal, type = "colour")
+  }
+  if (is_pal(pal)) {
+    return(pal)
+  }
+  cli::cli_abort("Failed to interpret {name} as palette.")
+}
+
+#' @export
+#' @rdname get_palette
+set_palette <- function(name, palette, warn_conflict = TRUE) {
+  name <- tolower(name)
+  if (!is_function(palette) && !is_character(palette)) {
+    cli::cli_abort(
+      "The {.arg palette} argument must be a {.cls function} or \\
+      {.cls character} vector."
+    )
+  }
+  if (warn_conflict & exists(name, envir = .known_palettes)) {
+    cli::cli_warn("Overwriting pre-existing {.val {name}} palette.")
+  }
+  env_bind(.known_palettes, !!name := palette)
+  invisible(NULL)
+}
+
+#' @export
+#' @rdname get_palette
+palette_names <- function() {
+  names(.known_palettes)
+}
+
+#' @export
+#' @rdname get_palette
+reset_palettes <- function() {
+  env_unbind(.known_palettes, palette_names())
+  init_palettes()
+}
+
+init_palettes <- function() {
+  register_hcl_pals()
+  register_base_pals()
+  register_viridis_pals()
+  register_brewer_pals()
+  register_dichromat_pals()
+  set_palette("grey", pal_grey, warn_conflict = FALSE)
+  set_palette("hue", pal_hue, warn_conflict = FALSE)
+}
+
+on_load(init_palettes())
+
+register_hcl_pals <- function(n = 31) {
+  names <- grDevices::hcl.pals()
+  for (name in names) {
+    fun <- colour_ramp(grDevices::hcl.colors(n, palette = name))
+    set_palette(name, fun, warn_conflict = FALSE)
+  }
+  invisible(NULL)
+}
+
+register_base_pals <- function() {
+  if (getRversion() < "4.0.0") {
+    return(invisible(NULL))
+  }
+  names   <- utils::getFromNamespace("palette.pals",   "grDevices")()
+  palette <- utils::getFromNamespace("palette.colors", "grDevices")
+  for (name in names) {
+    fun <- manual_pal(palette(palette = name), type = "colour")
+    set_palette(name, fun, warn_conflict = FALSE)
+  }
+  invisible(NULL)
+}
+
+register_viridis_pals <- function() {
+  names <- c("magma", "inferno", "plasma", "viridis",
+             "cividis", "rocket", "mako", "turbo")
+  for (name in names) {
+    fun <- pal_viridis(option = name)
+    set_palette(name, fun, warn_conflict = FALSE)
+  }
+  invisible(NULL)
+}
+
+register_brewer_pals <- function() {
+  names <- rownames(RColorBrewer::brewer.pal.info)
+  for (name in names) {
+    fun <- pal_brewer(palette = name)
+    set_palette(name, fun, warn_conflict = FALSE)
+  }
+  invisible(NULL)
+}
+
+register_dichromat_pals <- function() {
+  if (!is_installed("dichromat")) {
+    return(invisible(NULL))
+  }
+  names <- names(dichromat::colorschemes)
+  for (name in names) {
+    fun <- manual_pal(dichromat::colorschemes[[name]], type = "colour")
+    set_palette(name, fun, warn_conflict = FALSE)
+  }
+  invisible(NULL)
+}
diff --git a/R/range.R b/R/range.R
index e2aa3476..f74ec500 100644
--- a/R/range.R
+++ b/R/range.R
@@ -18,9 +18,10 @@ DiscreteRange <- R6::R6Class(
   inherit = Range,
   list(
     factor = NULL,
-    train = function(x, drop = FALSE, na.rm = FALSE) {
+    train = function(x, drop = FALSE, na.rm = FALSE, call = caller_env()) {
       self$factor <- self$factor %||% is.factor(x)
-      self$range <- train_discrete(x, self$range, drop, na.rm, self$factor)
+      self$range <- train_discrete(x, self$range, drop, na.rm,
+                                   self$factor, call = call)
     },
     reset = function() {
       self$range <- NULL
@@ -35,7 +36,9 @@ ContinuousRange <- R6::R6Class(
   "ContinuousRange",
   inherit = Range,
   list(
-    train = function(x) self$range <- train_continuous(x, self$range),
+    train = function(x, call = caller_env()) {
+      self$range <- train_continuous(x, self$range, call = call)
+    },
     reset = function() self$range <- NULL
   )
 )
diff --git a/R/scale-continuous.R b/R/scale-continuous.R
index 102b325f..b5095fbe 100644
--- a/R/scale-continuous.R
+++ b/R/scale-continuous.R
@@ -37,13 +37,18 @@ cscale <- function(x, palette, na.value = NA_real_, trans = transform_identity()
 #'
 #' @inheritParams train_discrete
 #' @export
-train_continuous <- function(new, existing = NULL) {
+train_continuous <- function(new, existing = NULL, call = caller_env()) {
   if (is.null(new)) {
     return(existing)
   }
 
   if (is.factor(new) || !typeof(new) %in% c("integer", "double")) {
-    cli::cli_abort("Discrete value supplied to a continuous scale")
+    example <- unique(new)
+    example <- example[seq_len(pmin(length(example), 5))]
+    cli::cli_abort(c(
+      "Discrete value supplied to a continuous scale.",
+      i = "Example values: {.and {.val {example}}}."
+    ), call = call)
   }
 
   # Needs casting to numeric because some `new` vectors can misbehave when
diff --git a/R/scale-discrete.R b/R/scale-discrete.R
index 3e95e13b..8957583a 100644
--- a/R/scale-discrete.R
+++ b/R/scale-discrete.R
@@ -25,14 +25,20 @@ is.discrete <- function(x) {
 #' @param drop `TRUE`, will drop factor levels not associated with data
 #' @param na.rm If `TRUE`, will remove missing values
 #' @param fct Treat `existing` as if it came from a factor (ie. don't sort the range)
+#' @param call A call to display in error messages
 #' @export
-train_discrete <- function(new, existing = NULL, drop = FALSE, na.rm = FALSE, fct = NA) {
+train_discrete <- function(new, existing = NULL, drop = FALSE,
+                           na.rm = FALSE, fct = NA, call = caller_env()) {
   if (is.null(new)) {
     return(existing)
   }
-
   if (!is.discrete(new)) {
-    cli::cli_abort("Continuous value supplied to a discrete scale")
+    example <- unique(new)
+    example <- example[seq_len(pmin(length(example), 5))]
+    cli::cli_abort(c(
+      "Continuous value supplied to a discrete scale.",
+      i = "Example values: {.and {.val {example}}}."
+    ), call = call)
   }
   discrete_range(existing, new, drop = drop, na.rm = na.rm, fct = fct)
 }
diff --git a/R/transform-numeric.R b/R/transform-numeric.R
index 641b7f53..0b55a1e1 100644
--- a/R/transform-numeric.R
+++ b/R/transform-numeric.R
@@ -67,16 +67,16 @@ asinh_trans <- transform_asinh
 #' Box-Cox to also work with negative values.
 #'
 #' The Box-Cox power transformation (type 1) requires strictly positive values and
-#' takes the following form for `y > 0`:
+#' takes the following form for \eqn{\lambda > 0}:
 #' \deqn{y^{(\lambda)} = \frac{y^\lambda - 1}{\lambda}}{y^(\lambda) = (y^\lambda - 1)/\lambda}
-#' When `y = 0`, the natural log transform is used.
+#' When \eqn{\lambda = 0}, the natural log transform is used.
 #'
 #' The modulus transformation implements a generalisation of the Box-Cox
 #' transformation that works for data with both positive and negative values.
-#' The equation takes the following forms, when `y != 0` :
+#' The equation takes the following forms, when \eqn{\lambda \neq 0} :
 #' \deqn{y^{(\lambda)} = sign(y) * \frac{(|y| + 1)^\lambda - 1}{\lambda}}{
 #' y^(\lambda) = sign(y)*((|y|+1)^\lambda - 1)/\lambda}
-#' and when `y = 0`: \deqn{y^{(\lambda)} =  sign(y) * \ln(|y| + 1)}{
+#' and when \eqn{\lambda = 0}: \deqn{y^{(\lambda)} =  sign(y) * \ln(|y| + 1)}{
 #' y^(\lambda) = sign(y) * ln(|y| + 1)}
 #'
 #' @param p Transformation exponent, \eqn{\lambda}.
@@ -255,7 +255,8 @@ transform_exp <- function(base = exp(1)) {
     function(x) base^x,
     function(x) log(x, base = base),
     d_transform = function(x) base^x * log(base),
-    d_inverse = function(x) 1 / x / log(base)
+    d_inverse = function(x) 1 / x / log(base),
+    breaks = breaks_exp(),
   )
 }
 
diff --git a/R/utils.R b/R/utils.R
index fc80a5a4..25cfa505 100644
--- a/R/utils.R
+++ b/R/utils.R
@@ -73,11 +73,7 @@ recycle_common <- function(..., size = NULL, call = caller_env()) {
   n <- setdiff(n, 1L)
   ns <- length(n)
 
-  if (ns == 0) { # All have length 1
-    if (is.null(size)) {
-      return(xs)
-    }
-  } else if (ns == 1) {
+  if (ns == 1) {
     if (is.null(size)) {
       size <- n
     } else if (n != size) {
@@ -100,3 +96,7 @@ recycle_common <- function(..., size = NULL, call = caller_env()) {
   x[to_recycle] <- lapply(x[to_recycle], rep_len, length.out = size)
   x
 }
+
+.onLoad <- function(lib, pkg) {
+  run_on_load()
+}
diff --git a/_pkgdown.yml b/_pkgdown.yml
index 3e26d4b2..12c48200 100644
--- a/_pkgdown.yml
+++ b/_pkgdown.yml
@@ -19,6 +19,7 @@ reference:
   contents:
   - starts_with("label_")
   - matches("format")
+  - number_options
 
 - title:  Axis breaks
   desc: >
@@ -27,6 +28,7 @@ reference:
   contents:
   - starts_with("breaks_")
   - minor_breaks_width
+  - minor_breaks_log
 
 - title: "Bounds: ranges & rescaling"
   desc: >
@@ -68,6 +70,7 @@ reference:
   - contains("col")
   - muted
   - alpha
+  - get_palette
 
 - title: Non-colour palette functions
   desc: >
diff --git a/man/breaks_exp.Rd b/man/breaks_exp.Rd
new file mode 100644
index 00000000..536b4412
--- /dev/null
+++ b/man/breaks_exp.Rd
@@ -0,0 +1,25 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/breaks.R
+\name{breaks_exp}
+\alias{breaks_exp}
+\title{Breaks for exponentially transformed data}
+\usage{
+breaks_exp(n = 5, ...)
+}
+\arguments{
+\item{n}{Desired number of breaks. You may get slightly more or fewer
+breaks that requested.}
+
+\item{...}{other arguments passed on to \code{\link[labeling:extended]{labeling::extended()}}}
+}
+\description{
+This breaks function typically labels zero and the last \code{n - 1} integers of a
+range if that range is large enough (currently: 3). For smaller ranges, it
+uses \code{\link[=breaks_extended]{breaks_extended()}}.
+}
+\examples{
+# Small range
+demo_continuous(c(100, 102), transform = "exp", breaks = breaks_exp())
+# Large range
+demo_continuous(c(0, 100), transform = "exp", breaks = breaks_exp(n = 4))
+}
diff --git a/man/comma.Rd b/man/comma.Rd
index 38f9b7f9..bb03854a 100644
--- a/man/comma.Rd
+++ b/man/comma.Rd
@@ -24,10 +24,10 @@ number_format(
   scale = 1,
   prefix = "",
   suffix = "",
-  big.mark = " ",
-  decimal.mark = ".",
-  style_positive = c("none", "plus", "space"),
-  style_negative = c("hyphen", "minus", "parens"),
+  big.mark = NULL,
+  decimal.mark = NULL,
+  style_positive = NULL,
+  style_negative = NULL,
   scale_cut = NULL,
   trim = TRUE,
   ...
@@ -65,10 +65,13 @@ processed so that \code{prefix = "$"} will yield (e.g.) \verb{-$1} and \verb{($1
 
 \item{suffix}{Additional text to display after the number.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{trim}{Logical, if \code{FALSE}, values are right-justified to a common
 width (see \code{\link[base:format]{base::format()}}).}
@@ -85,7 +88,10 @@ width (see \code{\link[base:format]{base::format()}}).}
 as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
 can ensure numbers remain properly aligned when they are left- or
 right-justified.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{style_negative}{A string that determines the style of negative numbers:
 \itemize{
@@ -94,7 +100,10 @@ right-justified.
 nicety that ensures \code{-} aligns with the horizontal bar of the
 the horizontal bar of \code{+}.
 \item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{scale_cut}{Named numeric vector that allows you to rescale large
 (or small) numbers and add a prefix. Built-in helpers include:
diff --git a/man/date_format.Rd b/man/date_format.Rd
index 8d3ab9e3..6de951f4 100644
--- a/man/date_format.Rd
+++ b/man/date_format.Rd
@@ -10,10 +10,10 @@ date_format(format = "\%Y-\%m-\%d", tz = "UTC", locale = NULL)
 time_format(format = "\%H:\%M:\%S", tz = "UTC", locale = NULL)
 }
 \arguments{
-\item{format}{For \code{date_format()} and \code{time_format()} a date/time format
+\item{format}{For \code{label_date()} and \code{label_time()} a date/time format
 string using standard POSIX specification.  See \code{\link[=strptime]{strptime()}} for details.
 
-For \code{date_short()} a character vector of length 4 giving the format
+For \code{label_date_short()} a character vector of length 4 giving the format
 components to use for year, month, day, and hour respectively.}
 
 \item{tz}{a time zone name, see \code{\link[=timezones]{timezones()}}. Defaults
diff --git a/man/dollar_format.Rd b/man/dollar_format.Rd
index fc65b2df..0791f2c6 100644
--- a/man/dollar_format.Rd
+++ b/man/dollar_format.Rd
@@ -23,10 +23,10 @@ dollar(
   x,
   accuracy = NULL,
   scale = 1,
-  prefix = "$",
-  suffix = "",
-  big.mark = ",",
-  decimal.mark = ".",
+  prefix = NULL,
+  suffix = NULL,
+  big.mark = NULL,
+  decimal.mark = NULL,
   trim = TRUE,
   largest_with_cents = 1e+05,
   negative_parens = deprecated(),
@@ -55,10 +55,13 @@ large.}
 
 \item{prefix, suffix}{Symbols to display before and after value.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{trim}{Logical, if \code{FALSE}, values are right-justified to a common
 width (see \code{\link[base:format]{base::format()}}).}
diff --git a/man/get_palette.Rd b/man/get_palette.Rd
new file mode 100644
index 00000000..e48def8e
--- /dev/null
+++ b/man/get_palette.Rd
@@ -0,0 +1,53 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/palette-registry.R
+\name{get_palette}
+\alias{get_palette}
+\alias{set_palette}
+\alias{palette_names}
+\alias{reset_palettes}
+\title{Known palettes}
+\usage{
+get_palette(name, ...)
+
+set_palette(name, palette, warn_conflict = TRUE)
+
+palette_names()
+
+reset_palettes()
+}
+\arguments{
+\item{name}{A string giving the palette name.}
+
+\item{...}{Additional arguments to pass to palette when it is a function
+but not a palette class function.}
+
+\item{palette}{A \link[=new_continuous_palette]{palette}, \code{function} or character
+vector.}
+
+\item{warn_conflict}{A boolean which if \code{TRUE} (default), warns when
+replacing a known palette.}
+}
+\value{
+The \code{get_palette()} function returns a palette. The \code{set_palette()}
+function is called for side effects and returns nothing.
+}
+\description{
+The scales packages keeps track of a set of palettes it considers 'known'.
+The benefit of a known palette is that it can be called by name in functions
+as \code{as_continuous_pal()} or \code{as_discrete_pal()}.
+}
+\examples{
+# Get one of the known palettes
+get_palette("hue")
+
+# Set a new custom palette
+cols <- c("palegreen", "deepskyblue", "magenta")
+set_palette("aurora", palette = cols)
+
+# Palette is now known
+"aurora" \%in\% palette_names()
+as_continuous_pal("aurora")
+
+# Resetting palettes
+reset_palettes()
+}
diff --git a/man/label_bytes.Rd b/man/label_bytes.Rd
index 02054433..266649bc 100644
--- a/man/label_bytes.Rd
+++ b/man/label_bytes.Rd
@@ -35,9 +35,12 @@ large.}
 applied to absolute value before \code{style_positive} and \code{style_negative} are
 processed so that \code{prefix = "$"} will yield (e.g.) \verb{-$1} and \verb{($1)}.}
     \item{\code{suffix}}{Additional text to display after the number.}
-    \item{\code{big.mark}}{Character used between every 3 digits to separate thousands.}
+    \item{\code{big.mark}}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{decimal.mark}}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_positive}}{A string that determines the style of positive numbers:
 \itemize{
 \item \code{"none"} (the default): no change, e.g. \code{1}.
@@ -46,7 +49,10 @@ decimal point.}
 as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
 can ensure numbers remain properly aligned when they are left- or
 right-justified.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_negative}}{A string that determines the style of negative numbers:
 \itemize{
 \item \code{"hyphen"} (the default): preceded by a standard hypen \code{-}, e.g. \code{-1}.
@@ -54,7 +60,10 @@ right-justified.
 nicety that ensures \code{-} aligns with the horizontal bar of the
 the horizontal bar of \code{+}.
 \item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{scale_cut}}{Named numeric vector that allows you to rescale large
 (or small) numbers and add a prefix. Built-in helpers include:
 \itemize{
diff --git a/man/label_currency.Rd b/man/label_currency.Rd
index a29feb17..c0118d92 100644
--- a/man/label_currency.Rd
+++ b/man/label_currency.Rd
@@ -7,10 +7,10 @@
 label_currency(
   accuracy = NULL,
   scale = 1,
-  prefix = "$",
-  suffix = "",
-  big.mark = ",",
-  decimal.mark = ".",
+  prefix = NULL,
+  suffix = NULL,
+  big.mark = NULL,
+  decimal.mark = NULL,
   trim = TRUE,
   largest_with_fractional = 1e+05,
   ...
@@ -29,10 +29,13 @@ large.}
 
 \item{prefix, suffix}{Symbols to display before and after value.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{trim}{Logical, if \code{FALSE}, values are right-justified to a common
 width (see \code{\link[base:format]{base::format()}}).}
@@ -48,7 +51,10 @@ width (see \code{\link[base:format]{base::format()}}).}
 as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
 can ensure numbers remain properly aligned when they are left- or
 right-justified.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_negative}}{A string that determines the style of negative numbers:
 \itemize{
 \item \code{"hyphen"} (the default): preceded by a standard hypen \code{-}, e.g. \code{-1}.
@@ -56,7 +62,10 @@ right-justified.
 nicety that ensures \code{-} aligns with the horizontal bar of the
 the horizontal bar of \code{+}.
 \item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{scale_cut}}{Named numeric vector that allows you to rescale large
 (or small) numbers and add a prefix. Built-in helpers include:
 \itemize{
diff --git a/man/label_date.Rd b/man/label_date.Rd
index e713f019..928f8cd5 100644
--- a/man/label_date.Rd
+++ b/man/label_date.Rd
@@ -24,10 +24,10 @@ label_timespan(
 )
 }
 \arguments{
-\item{format}{For \code{date_format()} and \code{time_format()} a date/time format
+\item{format}{For \code{label_date()} and \code{label_time()} a date/time format
 string using standard POSIX specification.  See \code{\link[=strptime]{strptime()}} for details.
 
-For \code{date_short()} a character vector of length 4 giving the format
+For \code{label_date_short()} a character vector of length 4 giving the format
 components to use for year, month, day, and hour respectively.}
 
 \item{tz}{a time zone name, see \code{\link[=timezones]{timezones()}}. Defaults
@@ -63,9 +63,12 @@ large.}
 applied to absolute value before \code{style_positive} and \code{style_negative} are
 processed so that \code{prefix = "$"} will yield (e.g.) \verb{-$1} and \verb{($1)}.}
     \item{\code{suffix}}{Additional text to display after the number.}
-    \item{\code{big.mark}}{Character used between every 3 digits to separate thousands.}
+    \item{\code{big.mark}}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{decimal.mark}}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_positive}}{A string that determines the style of positive numbers:
 \itemize{
 \item \code{"none"} (the default): no change, e.g. \code{1}.
@@ -74,7 +77,10 @@ decimal point.}
 as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
 can ensure numbers remain properly aligned when they are left- or
 right-justified.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_negative}}{A string that determines the style of negative numbers:
 \itemize{
 \item \code{"hyphen"} (the default): preceded by a standard hypen \code{-}, e.g. \code{-1}.
@@ -82,7 +88,10 @@ right-justified.
 nicety that ensures \code{-} aligns with the horizontal bar of the
 the horizontal bar of \code{+}.
 \item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{trim}}{Logical, if \code{FALSE}, values are right-justified to a common
 width (see \code{\link[base:format]{base::format()}}).}
   }}
@@ -116,9 +125,9 @@ date_range <- function(start, days) {
 
 two_months <- date_range("2020-05-01", 60)
 demo_datetime(two_months)
-demo_datetime(two_months, labels = date_format("\%m/\%d"))
-demo_datetime(two_months, labels = date_format("\%e \%b", locale = "fr"))
-demo_datetime(two_months, labels = date_format("\%e \%B", locale = "es"))
+demo_datetime(two_months, labels = label_date("\%m/\%d"))
+demo_datetime(two_months, labels = label_date("\%e \%b", locale = "fr"))
+demo_datetime(two_months, labels = label_date("\%e \%B", locale = "es"))
 # ggplot2 provides a short-hand:
 demo_datetime(two_months, date_labels = "\%m/\%d")
 
diff --git a/man/label_log.Rd b/man/label_log.Rd
index daf576fb..d0dee6ab 100644
--- a/man/label_log.Rd
+++ b/man/label_log.Rd
@@ -2,15 +2,26 @@
 % Please edit documentation in R/label-log.R
 \name{label_log}
 \alias{label_log}
+\alias{format_log}
 \title{Label numbers in log format (10^3, 10^6, etc)}
 \usage{
-label_log(base = 10, digits = 3)
+label_log(base = 10, digits = 3, signed = NULL)
+
+format_log(x, base = 10, signed = NULL, ...)
 }
 \arguments{
 \item{base}{Base of logarithm to use}
 
 \item{digits}{Number of significant digits to show for the exponent. Argument
 is passed on to \code{\link[base:format]{base::format()}}.}
+
+\item{signed}{Should a \code{+} or \code{-} be displayed as a prefix? The
+default, \code{NULL}, displays signs if there are zeroes or negative numbers
+present.}
+
+\item{x}{A numeric vector to format}
+
+\item{...}{Passed on to \code{format()}.}
 }
 \value{
 All \code{label_()} functions return a "labelling" function, i.e. a function that
@@ -23,11 +34,14 @@ they work similarly for all scales, including those that generate legends
 rather than axes.
 }
 \description{
-\code{label_log()} displays numbers as base^exponent, using superscript formatting.
+\code{label_log()} and \code{format_log()} display numbers as base^exponent, using
+superscript formatting. \code{label_log()} returns expressions suitable for
+labelling in scales, whereas \code{format_log()} returns deparsed text.
 }
 \examples{
 demo_log10(c(1, 1e5), labels = label_log())
 demo_log10(c(1, 1e5), breaks = breaks_log(base = 2), labels = label_log(base = 2))
+format_log(c(0.1, 1, 10))
 }
 \seealso{
 \code{\link[=breaks_log]{breaks_log()}} for the related breaks algorithm.
diff --git a/man/label_number.Rd b/man/label_number.Rd
index cb91f2a9..322f20db 100644
--- a/man/label_number.Rd
+++ b/man/label_number.Rd
@@ -10,10 +10,10 @@ label_number(
   scale = 1,
   prefix = "",
   suffix = "",
-  big.mark = " ",
-  decimal.mark = ".",
-  style_positive = c("none", "plus", "space"),
-  style_negative = c("hyphen", "minus", "parens"),
+  big.mark = NULL,
+  decimal.mark = NULL,
+  style_positive = NULL,
+  style_negative = NULL,
   scale_cut = NULL,
   trim = TRUE,
   ...
@@ -49,10 +49,13 @@ processed so that \code{prefix = "$"} will yield (e.g.) \verb{-$1} and \verb{($1
 
 \item{suffix}{Additional text to display after the number.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{style_positive}{A string that determines the style of positive numbers:
 \itemize{
@@ -62,7 +65,10 @@ decimal point.}
 as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
 can ensure numbers remain properly aligned when they are left- or
 right-justified.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{style_negative}{A string that determines the style of negative numbers:
 \itemize{
@@ -71,7 +77,10 @@ right-justified.
 nicety that ensures \code{-} aligns with the horizontal bar of the
 the horizontal bar of \code{+}.
 \item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{scale_cut}{Named numeric vector that allows you to rescale large
 (or small) numbers and add a prefix. Built-in helpers include:
diff --git a/man/label_number_si.Rd b/man/label_number_si.Rd
index fd8c8ece..c3a9eeaf 100644
--- a/man/label_number_si.Rd
+++ b/man/label_number_si.Rd
@@ -28,9 +28,12 @@ prefix.}
     \item{\code{prefix}}{Additional text to display before the number. The suffix is
 applied to absolute value before \code{style_positive} and \code{style_negative} are
 processed so that \code{prefix = "$"} will yield (e.g.) \verb{-$1} and \verb{($1)}.}
-    \item{\code{big.mark}}{Character used between every 3 digits to separate thousands.}
+    \item{\code{big.mark}}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{decimal.mark}}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_positive}}{A string that determines the style of positive numbers:
 \itemize{
 \item \code{"none"} (the default): no change, e.g. \code{1}.
@@ -39,7 +42,10 @@ decimal point.}
 as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
 can ensure numbers remain properly aligned when they are left- or
 right-justified.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_negative}}{A string that determines the style of negative numbers:
 \itemize{
 \item \code{"hyphen"} (the default): preceded by a standard hypen \code{-}, e.g. \code{-1}.
@@ -47,7 +53,10 @@ right-justified.
 nicety that ensures \code{-} aligns with the horizontal bar of the
 the horizontal bar of \code{+}.
 \item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{scale_cut}}{Named numeric vector that allows you to rescale large
 (or small) numbers and add a prefix. Built-in helpers include:
 \itemize{
diff --git a/man/label_ordinal.Rd b/man/label_ordinal.Rd
index 471c0ebd..77da0587 100644
--- a/man/label_ordinal.Rd
+++ b/man/label_ordinal.Rd
@@ -7,13 +7,7 @@
 \alias{ordinal_spanish}
 \title{Label ordinal numbers (1st, 2nd, 3rd, etc)}
 \usage{
-label_ordinal(
-  prefix = "",
-  suffix = "",
-  big.mark = " ",
-  rules = ordinal_english(),
-  ...
-)
+label_ordinal(prefix = "", suffix = "", big.mark = NULL, rules = NULL, ...)
 
 ordinal_english()
 
@@ -24,7 +18,9 @@ ordinal_spanish()
 \arguments{
 \item{prefix, suffix}{Symbols to display before and after value.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{rules}{Named list of regular expressions, matched in order.
 Name gives suffix, and value specifies which numbers to match.}
@@ -42,7 +38,8 @@ Applied to rescaled data.}
 formatting. This is useful if the underlying data is very small or very
 large.}
     \item{\code{decimal.mark}}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_positive}}{A string that determines the style of positive numbers:
 \itemize{
 \item \code{"none"} (the default): no change, e.g. \code{1}.
@@ -51,7 +48,10 @@ decimal point.}
 as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
 can ensure numbers remain properly aligned when they are left- or
 right-justified.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_negative}}{A string that determines the style of negative numbers:
 \itemize{
 \item \code{"hyphen"} (the default): preceded by a standard hypen \code{-}, e.g. \code{-1}.
@@ -59,7 +59,10 @@ right-justified.
 nicety that ensures \code{-} aligns with the horizontal bar of the
 the horizontal bar of \code{+}.
 \item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{scale_cut}}{Named numeric vector that allows you to rescale large
 (or small) numbers and add a prefix. Built-in helpers include:
 \itemize{
diff --git a/man/label_percent.Rd b/man/label_percent.Rd
index a705538c..4fe707a8 100644
--- a/man/label_percent.Rd
+++ b/man/label_percent.Rd
@@ -9,8 +9,8 @@ label_percent(
   scale = 100,
   prefix = "",
   suffix = "\%",
-  big.mark = " ",
-  decimal.mark = ".",
+  big.mark = NULL,
+  decimal.mark = NULL,
   trim = TRUE,
   ...
 )
@@ -33,10 +33,13 @@ processed so that \code{prefix = "$"} will yield (e.g.) \verb{-$1} and \verb{($1
 
 \item{suffix}{Additional text to display after the number.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{trim}{Logical, if \code{FALSE}, values are right-justified to a common
 width (see \code{\link[base:format]{base::format()}}).}
@@ -52,7 +55,10 @@ width (see \code{\link[base:format]{base::format()}}).}
 as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
 can ensure numbers remain properly aligned when they are left- or
 right-justified.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{style_negative}}{A string that determines the style of negative numbers:
 \itemize{
 \item \code{"hyphen"} (the default): preceded by a standard hypen \code{-}, e.g. \code{-1}.
@@ -60,7 +66,10 @@ right-justified.
 nicety that ensures \code{-} aligns with the horizontal bar of the
 the horizontal bar of \code{+}.
 \item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
     \item{\code{scale_cut}}{Named numeric vector that allows you to rescale large
 (or small) numbers and add a prefix. Built-in helpers include:
 \itemize{
diff --git a/man/label_pvalue.Rd b/man/label_pvalue.Rd
index ae9e2b83..f38d49f6 100644
--- a/man/label_pvalue.Rd
+++ b/man/label_pvalue.Rd
@@ -6,7 +6,7 @@
 \usage{
 label_pvalue(
   accuracy = 0.001,
-  decimal.mark = ".",
+  decimal.mark = NULL,
   prefix = NULL,
   add_p = FALSE
 )
@@ -20,11 +20,12 @@ difference between adjacent values.
 Applied to rescaled data.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{prefix}{A character vector of length 3 giving the prefixes to
-put in front of numbers. The default values are \code{c("<", "", ">")}
-if \code{add_p} is \code{TRUE} and \code{c("p<", "p=", "p>")} if \code{FALSE}.}
+put in front of numbers. The default values are \code{c("p<", "p=", "p>")}
+if \code{add_p} is \code{TRUE} and \code{c("<", "", ">")} if \code{FALSE}.}
 
 \item{add_p}{Add "p=" before the value?}
 }
diff --git a/man/label_scientific.Rd b/man/label_scientific.Rd
index 97db3367..b28a8608 100644
--- a/man/label_scientific.Rd
+++ b/man/label_scientific.Rd
@@ -9,7 +9,7 @@ label_scientific(
   scale = 1,
   prefix = "",
   suffix = "",
-  decimal.mark = ".",
+  decimal.mark = NULL,
   trim = TRUE,
   ...
 )
@@ -24,7 +24,8 @@ large.}
 \item{prefix, suffix}{Symbols to display before and after value.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{trim}{Logical, if \code{FALSE}, values are right-justified to a common
 width (see \code{\link[base:format]{base::format()}}).}
diff --git a/man/minor_breaks_log.Rd b/man/minor_breaks_log.Rd
new file mode 100644
index 00000000..1efb1da2
--- /dev/null
+++ b/man/minor_breaks_log.Rd
@@ -0,0 +1,34 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/breaks-log.R
+\name{minor_breaks_log}
+\alias{minor_breaks_log}
+\title{Minor breaks for log-10 axes}
+\usage{
+minor_breaks_log(detail = NULL, smallest = NULL)
+}
+\arguments{
+\item{detail}{Any of \code{1}, \code{5} and \code{10} to mark multiples of
+powers, multiples of 5 of powers or just powers respectively.}
+
+\item{smallest}{Smallest absolute value to mark when the range includes
+negative numbers.}
+}
+\value{
+A function to generate minor ticks.
+}
+\description{
+This break function is designed to mark every power, multiples of 5 and/or 1
+of that power for base 10.
+}
+\examples{
+# Standard usage with log10 scale
+demo_log10(c(1, 1e10), minor_breaks = minor_breaks_log())
+# Increasing detail over many powers
+demo_log10(c(1, 1e10), minor_breaks = minor_breaks_log(detail = 1))
+# Adjusting until where to draw minor breaks
+demo_continuous(
+  c(-1000, 1000),
+  transform = asinh_trans(),
+  minor_breaks = minor_breaks_log(smallest = 1)
+)
+}
diff --git a/man/number.Rd b/man/number.Rd
index 2ac5a42c..57696e96 100644
--- a/man/number.Rd
+++ b/man/number.Rd
@@ -14,10 +14,10 @@ number(
   scale = 1,
   prefix = "",
   suffix = "",
-  big.mark = " ",
-  decimal.mark = ".",
-  style_positive = c("none", "plus", "space"),
-  style_negative = c("hyphen", "minus", "parens"),
+  big.mark = NULL,
+  decimal.mark = NULL,
+  style_positive = NULL,
+  style_negative = NULL,
   scale_cut = NULL,
   trim = TRUE,
   ...
@@ -49,10 +49,13 @@ processed so that \code{prefix = "$"} will yield (e.g.) \verb{-$1} and \verb{($1
 
 \item{suffix}{Additional text to display after the number.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{style_positive}{A string that determines the style of positive numbers:
 \itemize{
@@ -62,7 +65,10 @@ decimal point.}
 as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
 can ensure numbers remain properly aligned when they are left- or
 right-justified.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{style_negative}{A string that determines the style of negative numbers:
 \itemize{
@@ -71,7 +77,10 @@ right-justified.
 nicety that ensures \code{-} aligns with the horizontal bar of the
 the horizontal bar of \code{+}.
 \item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
-}}
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{scale_cut}{Named numeric vector that allows you to rescale large
 (or small) numbers and add a prefix. Built-in helpers include:
diff --git a/man/number_options.Rd b/man/number_options.Rd
new file mode 100644
index 00000000..d9c81aa5
--- /dev/null
+++ b/man/number_options.Rd
@@ -0,0 +1,79 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/label-number.R
+\name{number_options}
+\alias{number_options}
+\title{Number options}
+\usage{
+number_options(
+  decimal.mark = ".",
+  big.mark = " ",
+  style_positive = c("none", "plus", "space"),
+  style_negative = c("hyphen", "minus", "parens"),
+  currency.prefix = "$",
+  currency.suffix = "",
+  currency.decimal.mark = decimal.mark,
+  currency.big.mark = setdiff(c(".", ","), currency.decimal.mark)[1],
+  ordinal.rules = ordinal_english()
+)
+}
+\arguments{
+\item{decimal.mark}{The character to be used to indicate the numeric
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
+
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
+
+\item{style_positive}{A string that determines the style of positive numbers:
+\itemize{
+\item \code{"none"} (the default): no change, e.g. \code{1}.
+\item \code{"plus"}: preceded by \code{+}, e.g. \code{+1}.
+\item \code{"space"}: preceded by a Unicode "figure space", i.e., a space equally
+as wide as a number or \code{+}. Compared to \code{"none"}, adding a figure space
+can ensure numbers remain properly aligned when they are left- or
+right-justified.
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
+
+\item{style_negative}{A string that determines the style of negative numbers:
+\itemize{
+\item \code{"hyphen"} (the default): preceded by a standard hypen \code{-}, e.g. \code{-1}.
+\item \code{"minus"}, uses a proper Unicode minus symbol. This is a typographical
+nicety that ensures \code{-} aligns with the horizontal bar of the
+the horizontal bar of \code{+}.
+\item \code{"parens"}, wrapped in parentheses, e.g. \code{(1)}.
+}
+
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
+
+\item{currency.prefix, currency.suffix, currency.decimal.mark, currency.big.mark}{Settings for \code{\link[=label_currency]{label_currency()}} passed on without the \code{currency.}-prefix.}
+
+\item{ordinal.rules}{Setting for \code{\link[=label_ordinal]{label_ordinal()}} passed on without the
+\code{ordinal.}-prefix.}
+}
+\value{
+The old options invisibly
+}
+\description{
+Control the settings for formatting numbers globally.
+}
+\examples{
+# Default number formatting
+x <- c(0.1, 1, 1000)
+label_number()(x)
+
+# Now again with new options set
+number_options(style_positive = "plus", decimal.mark = ",")
+label_number()(x)
+
+# The options are the argument names with a 'scales.'-prefix
+options("scales.style_positive")
+
+# Resetting the options to their defaults
+number_options()
+label_number()(x)
+}
diff --git a/man/ordinal_format.Rd b/man/ordinal_format.Rd
index c2a692a3..1de83ed9 100644
--- a/man/ordinal_format.Rd
+++ b/man/ordinal_format.Rd
@@ -5,27 +5,16 @@
 \alias{ordinal}
 \title{Superseded interface to \code{label_ordinal()}}
 \usage{
-ordinal_format(
-  prefix = "",
-  suffix = "",
-  big.mark = " ",
-  rules = ordinal_english(),
-  ...
-)
+ordinal_format(prefix = "", suffix = "", big.mark = NULL, rules = NULL, ...)
 
-ordinal(
-  x,
-  prefix = "",
-  suffix = "",
-  big.mark = " ",
-  rules = ordinal_english(),
-  ...
-)
+ordinal(x, prefix = "", suffix = "", big.mark = " ", rules = NULL, ...)
 }
 \arguments{
 \item{prefix, suffix}{Symbols to display before and after value.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{rules}{Named list of regular expressions, matched in order.
 Name gives suffix, and value specifies which numbers to match.}
diff --git a/man/percent_format.Rd b/man/percent_format.Rd
index aad6ad30..26120995 100644
--- a/man/percent_format.Rd
+++ b/man/percent_format.Rd
@@ -10,8 +10,8 @@ percent_format(
   scale = 100,
   prefix = "",
   suffix = "\%",
-  big.mark = " ",
-  decimal.mark = ".",
+  big.mark = NULL,
+  decimal.mark = NULL,
   trim = TRUE,
   ...
 )
@@ -22,8 +22,8 @@ percent(
   scale = 100,
   prefix = "",
   suffix = "\%",
-  big.mark = " ",
-  decimal.mark = ".",
+  big.mark = NULL,
+  decimal.mark = NULL,
   trim = TRUE,
   ...
 )
@@ -46,10 +46,13 @@ processed so that \code{prefix = "$"} will yield (e.g.) \verb{-$1} and \verb{($1
 
 \item{suffix}{Additional text to display after the number.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{trim}{Logical, if \code{FALSE}, values are right-justified to a common
 width (see \code{\link[base:format]{base::format()}}).}
diff --git a/man/pvalue_format.Rd b/man/pvalue_format.Rd
index 71e4436a..a6be62a3 100644
--- a/man/pvalue_format.Rd
+++ b/man/pvalue_format.Rd
@@ -7,12 +7,12 @@
 \usage{
 pvalue_format(
   accuracy = 0.001,
-  decimal.mark = ".",
+  decimal.mark = NULL,
   prefix = NULL,
   add_p = FALSE
 )
 
-pvalue(x, accuracy = 0.001, decimal.mark = ".", prefix = NULL, add_p = FALSE)
+pvalue(x, accuracy = 0.001, decimal.mark = NULL, prefix = NULL, add_p = FALSE)
 }
 \arguments{
 \item{accuracy}{A number to round to. Use (e.g.) \code{0.01} to show 2 decimal
@@ -23,11 +23,12 @@ difference between adjacent values.
 Applied to rescaled data.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{prefix}{A character vector of length 3 giving the prefixes to
-put in front of numbers. The default values are \code{c("<", "", ">")}
-if \code{add_p} is \code{TRUE} and \code{c("p<", "p=", "p>")} if \code{FALSE}.}
+put in front of numbers. The default values are \code{c("p<", "p=", "p>")}
+if \code{add_p} is \code{TRUE} and \code{c("<", "", ">")} if \code{FALSE}.}
 
 \item{add_p}{Add "p=" before the value?}
 }
diff --git a/man/scientific_format.Rd b/man/scientific_format.Rd
index dfa483d9..e0b2e715 100644
--- a/man/scientific_format.Rd
+++ b/man/scientific_format.Rd
@@ -10,7 +10,7 @@ scientific_format(
   scale = 1,
   prefix = "",
   suffix = "",
-  decimal.mark = ".",
+  decimal.mark = NULL,
   trim = TRUE,
   ...
 )
@@ -21,7 +21,7 @@ scientific(
   scale = 1,
   prefix = "",
   suffix = "",
-  decimal.mark = ".",
+  decimal.mark = NULL,
   trim = TRUE,
   ...
 )
@@ -36,7 +36,8 @@ large.}
 \item{prefix, suffix}{Symbols to display before and after value.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{trim}{Logical, if \code{FALSE}, values are right-justified to a common
 width (see \code{\link[base:format]{base::format()}}).}
diff --git a/man/train_continuous.Rd b/man/train_continuous.Rd
index ece41995..16d62955 100644
--- a/man/train_continuous.Rd
+++ b/man/train_continuous.Rd
@@ -4,12 +4,14 @@
 \alias{train_continuous}
 \title{Train (update) a continuous scale}
 \usage{
-train_continuous(new, existing = NULL)
+train_continuous(new, existing = NULL, call = caller_env())
 }
 \arguments{
 \item{new}{New data to add to scale}
 
 \item{existing}{Optional existing scale to update}
+
+\item{call}{A call to display in error messages}
 }
 \description{
 Strips attributes and always returns a numeric vector
diff --git a/man/train_discrete.Rd b/man/train_discrete.Rd
index 9e717429..cbeb1ef4 100644
--- a/man/train_discrete.Rd
+++ b/man/train_discrete.Rd
@@ -4,7 +4,14 @@
 \alias{train_discrete}
 \title{Train (update) a discrete scale}
 \usage{
-train_discrete(new, existing = NULL, drop = FALSE, na.rm = FALSE, fct = NA)
+train_discrete(
+  new,
+  existing = NULL,
+  drop = FALSE,
+  na.rm = FALSE,
+  fct = NA,
+  call = caller_env()
+)
 }
 \arguments{
 \item{new}{New data to add to scale}
@@ -16,6 +23,8 @@ train_discrete(new, existing = NULL, drop = FALSE, na.rm = FALSE, fct = NA)
 \item{na.rm}{If \code{TRUE}, will remove missing values}
 
 \item{fct}{Treat \code{existing} as if it came from a factor (ie. don't sort the range)}
+
+\item{call}{A call to display in error messages}
 }
 \description{
 Train (update) a discrete scale
diff --git a/man/transform_boxcox.Rd b/man/transform_boxcox.Rd
index cc6d8ea5..00d9ca28 100644
--- a/man/transform_boxcox.Rd
+++ b/man/transform_boxcox.Rd
@@ -29,16 +29,16 @@ Box-Cox to also work with negative values.
 }
 \details{
 The Box-Cox power transformation (type 1) requires strictly positive values and
-takes the following form for \code{y > 0}:
+takes the following form for \eqn{\lambda > 0}:
 \deqn{y^{(\lambda)} = \frac{y^\lambda - 1}{\lambda}}{y^(\lambda) = (y^\lambda - 1)/\lambda}
-When \code{y = 0}, the natural log transform is used.
+When \eqn{\lambda = 0}, the natural log transform is used.
 
 The modulus transformation implements a generalisation of the Box-Cox
 transformation that works for data with both positive and negative values.
-The equation takes the following forms, when \code{y != 0} :
+The equation takes the following forms, when \eqn{\lambda \neq 0} :
 \deqn{y^{(\lambda)} = sign(y) * \frac{(|y| + 1)^\lambda - 1}{\lambda}}{
 y^(\lambda) = sign(y)*((|y|+1)^\lambda - 1)/\lambda}
-and when \code{y = 0}: \deqn{y^{(\lambda)} =  sign(y) * \ln(|y| + 1)}{
+and when \eqn{\lambda = 0}: \deqn{y^{(\lambda)} =  sign(y) * \ln(|y| + 1)}{
 y^(\lambda) = sign(y) * ln(|y| + 1)}
 }
 \examples{
diff --git a/man/unit_format.Rd b/man/unit_format.Rd
index edbaad83..615a2411 100644
--- a/man/unit_format.Rd
+++ b/man/unit_format.Rd
@@ -11,8 +11,8 @@ unit_format(
   unit = "m",
   sep = " ",
   suffix = paste0(sep, unit),
-  big.mark = " ",
-  decimal.mark = ".",
+  big.mark = NULL,
+  decimal.mark = NULL,
   trim = TRUE,
   ...
 )
@@ -39,10 +39,13 @@ processed so that \code{prefix = "$"} will yield (e.g.) \verb{-$1} and \verb{($1
 
 \item{suffix}{Additional text to display after the number.}
 
-\item{big.mark}{Character used between every 3 digits to separate thousands.}
+\item{big.mark}{Character used between every 3 digits to separate thousands.
+The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{decimal.mark}{The character to be used to indicate the numeric
-decimal point.}
+decimal point.  The default (\code{NULL}) retrieves the setting from the
+\link[=number_options]{number options}.}
 
 \item{trim}{Logical, if \code{FALSE}, values are right-justified to a common
 width (see \code{\link[base:format]{base::format()}}).}
diff --git a/tests/testthat/_snaps/breaks-log.md b/tests/testthat/_snaps/breaks-log.md
new file mode 100644
index 00000000..aa7153eb
--- /dev/null
+++ b/tests/testthat/_snaps/breaks-log.md
@@ -0,0 +1,16 @@
+# minor_breaks_log rejects invalid arguments
+
+    Code
+      minor_breaks_log(7)
+    Condition
+      Error in `minor_breaks_log()`:
+      ! The `detail` argument must be one of 1, 5 or 10.
+
+---
+
+    Code
+      minor_breaks_log(smallest = 0)
+    Condition
+      Error in `minor_breaks_log()`:
+      ! The `smallest` argument must be a finite, positive, non-zero number.
+
diff --git a/tests/testthat/test-breaks-log.R b/tests/testthat/test-breaks-log.R
index be32793e..5cbecb8c 100644
--- a/tests/testthat/test-breaks-log.R
+++ b/tests/testthat/test-breaks-log.R
@@ -90,3 +90,26 @@ test_that("breaks_log with very small ranges fall back to extended_breaks", {
     extended_breaks(n = 5)(c(0.95, 3))
   ))
 })
+
+test_that("minor_breaks_log has correct amount of detail", {
+  range <- c(1, 10)
+
+  test <- minor_breaks_log(detail = 1)(range)
+  expect_true(all(1:10 %in% test))
+
+  test <- minor_breaks_log(detail = 5)(range)
+  expect_false(all(1:10 %in% test))
+  expect_true(all(c(1, 5, 10) %in% test))
+
+  test <- minor_breaks_log(detail = 10)(range)
+  expect_true(all(c(1, 10) %in% test))
+  expect_false(5 %in% test)
+
+  test <- minor_breaks_log(detail = 1)(c(-10, 10))
+  expect_true(all(-10:10 %in%  test))
+})
+
+test_that("minor_breaks_log rejects invalid arguments", {
+  expect_snapshot(minor_breaks_log(7), error = TRUE)
+  expect_snapshot(minor_breaks_log(smallest = 0), error = TRUE)
+})
diff --git a/tests/testthat/test-breaks.R b/tests/testthat/test-breaks.R
index 121b3400..0ceddca0 100644
--- a/tests/testthat/test-breaks.R
+++ b/tests/testthat/test-breaks.R
@@ -30,3 +30,21 @@ test_that("breaks_pretty() arguments are forcely evaluated on each call #81", {
   expect_equal(subfun1(1), subfuns[[1]](1))
   expect_equal(subfun2(1), subfuns[[2]](1))
 })
+
+test_that("breaks_pretty() returns input when given zero-width range (#446)", {
+  expect_equal(breaks_pretty()(c(1, 1)), 1)
+})
+
+test_that("exponential breaks give sensible values", {
+  x <- breaks_exp()(c(0, 2))
+  expect_equal(x, c(0, 0.5, 1, 1.5, 2))
+
+  x <- breaks_exp()(c(0, 5))
+  expect_equal(x, c(5, 4, 3, 2, 0))
+
+  x <- breaks_exp()(c(100, 102))
+  expect_equal(x, c(0, 0.5, 1, 1.5, 2) + 100)
+
+  x <- breaks_exp()(c(0, 100))
+  expect_equal(x, c(100, 99, 98, 97, 0))
+})
diff --git a/tests/testthat/test-label-log.R b/tests/testthat/test-label-log.R
index 0a555ae0..2aee4f49 100644
--- a/tests/testthat/test-label-log.R
+++ b/tests/testthat/test-label-log.R
@@ -5,4 +5,5 @@ test_that("label_log() returns expression", {
   expect_equal(label_log()(c(0.1, 10)), expression(10^-1, 10^1))
   expect_equal(label_log(base = 2)(8), expression(2^3))
   expect_equal(label_log(base = 2, digits = 3)(7), expression(2^2.81))
+  expect_equal(label_log(signed = TRUE)(c(-100, 100)), expression(-10^2, +10^2))
 })
diff --git a/tests/testthat/test-label-number.R b/tests/testthat/test-label-number.R
index 4a4cd490..d003816e 100644
--- a/tests/testthat/test-label-number.R
+++ b/tests/testthat/test-label-number.R
@@ -179,6 +179,11 @@ test_that("scale_cut prefers clean cuts", {
   x <- c(518400, 691200)
   # prefers days over week in second element
   expect_equal(number(x, scale_cut = cut_time_scale()), c("6d", "8d"))
+
+  # do not select off-scale breaks
+  x <- c(0, 500, 1500, 2000, 2500)
+  expect_equal(number(x, scale_cut = cut_short_scale()), c("0", "500", "1.5K", "2.0K", "2.5K"))
+
 })
 
 test_that("built-in functions return expected values", {
@@ -188,3 +193,49 @@ test_that("built-in functions return expected values", {
   expect_equal(number(1e9, scale_cut = cut_long_scale()), "1 000M")
   expect_equal(number(1e9, scale_cut = cut_si("m")), "1 Gm")
 })
+
+# options -----------------------------------------------------------------
+
+test_that("number_options are observed", {
+
+  number_options(decimal.mark = ",", big.mark = ".")
+  expect_equal(
+    label_number()(c(0.1, 10, 1e6)),
+    c("0,1", "10,0", "1.000.000,0")
+  )
+
+  number_options(style_positive = "plus", style_negative = "parens")
+  expect_equal(
+    label_number()(c(-0.1, 0, 1)),
+    c("(0.1)", "0.0", "+1.0")
+  )
+
+  number_options(
+    currency.prefix = "", currency.suffix = " GBP",
+    currency.decimal.mark = ",", currency.big.mark = "."
+  )
+  # Regular number are not affected
+  expect_equal(
+    label_number()(c(0.1, 10, 1e6)),
+    c("0.1", "10.0", "1 000 000.0")
+  )
+  # But currency is affected
+  expect_equal(
+    label_currency(accuracy = 0.1)(c(0.1, 10, 1e6)),
+    c("0,1 GBP", "10,0 GBP", "1.000.000,0 GBP")
+  )
+
+  number_options(ordinal.rules = ordinal_french(plural = TRUE))
+  expect_equal(
+    label_ordinal()(1:4),
+    c("1ers", "2es", "3es", "4es")
+  )
+
+  # Can be reset
+  number_options()
+  expect_equal(
+    label_ordinal()(1:4),
+    c("1st", "2nd", "3rd", "4th")
+  )
+})
+
diff --git a/tests/testthat/test-palette-registry.R b/tests/testthat/test-palette-registry.R
new file mode 100644
index 00000000..06e09074
--- /dev/null
+++ b/tests/testthat/test-palette-registry.R
@@ -0,0 +1,41 @@
+test_that("palette getters and setters work as intended", {
+
+  # Test that palettes have been populated in .onLoad
+  expect_in(c("hue", "grey"), palette_names())
+
+  # We cannot get unknown palettes
+  expect_error(get_palette("rgb"), "Unknown palette")
+
+  # We cannot set nonsense palettes
+  expect_error(
+    set_palette("foobar", list(a = 1:2, b = "A")),
+    "must be a"
+  )
+
+  # Test we can set custom palettes
+  colours <- c("red", "green", 'blue')
+  set_palette("rgb", palette = colours)
+  expect_in("rgb", palette_names())
+
+  # Test we can get custom palettes
+  pal <- get_palette("rgb")
+  expect_equal(pal(length(colours)), colours)
+
+  # Test we can reset palettes
+  reset_palettes()
+  expect_false("rgb" %in% palette_names())
+})
+
+test_that("as_continuous_pal and as_discrete_pal can retrieve known palettes", {
+
+  colours <- c("#FF0000", "#00FF00", '#0000FF')
+  set_palette("rgb", colours)
+
+  pal <- as_discrete_pal("rgb")
+  expect_equal(pal(length(colours)), colours)
+
+  pal <- as_continuous_pal("rgb")
+  expect_equal(pal(seq(0, 1, length.out = length(colours))), colours)
+
+  reset_palettes()
+})