diff --git a/R/read.R b/R/read.R
index 7717c07fa..84d1d95e2 100644
--- a/R/read.R
+++ b/R/read.R
@@ -309,7 +309,7 @@ wb_to_df <- function(
   keep_rows <- keep_rows[keep_rows %in% rnams]
 
   # reduce data to selected cases only
-  if (!is.null(cols) && !is.null(rows) && !missing(dims))
+  if (length(keep_rows) && length(keep_cols))
     cc <- cc[cc$row_r %in% keep_rows & cc$c_r %in% keep_cols, ]
 
   cc$val <- NA_character_
diff --git a/tests/testthat/test-wb_functions.R b/tests/testthat/test-wb_functions.R
index d717e03a3..c4352d96f 100644
--- a/tests/testthat/test-wb_functions.R
+++ b/tests/testthat/test-wb_functions.R
@@ -304,3 +304,14 @@ test_that("cols return order is correct", {
   expect_equal(exp, got)
 
 })
+
+test_that("missings cells are returned", {
+
+  wb <- wb_workbook()$add_worksheet()$add_data(x = "a", dims = "Z100")
+
+  expect_silent(got <- wb_to_df(wb, col_names = FALSE, rows = 1, cols = 1))
+  expect_silent(got <- wb_to_df(wb, col_names = FALSE, rows = c(1, 4, 10), cols = c(2, 5, 10)))
+
+  expect_silent(got <- wb_to_df(wb, col_names = FALSE, rows = c(101), cols = c(27)))
+
+})