white foxglove, Digitalis purpurea

Foxglove Falsehood

I know I read somewhere that the flower colour of a foxglove (Digitalis purpurea) plant can be predicted by the colour of the leaf stems (petioles). Foxgloves are biennials; they flower in their second year of life, and then they die. Their normal flower colour is a reddish purple. It’s fairly attractive, but not nearly as elegant as the white form. Fussy gardeners who prefer the white form would therefore want to know which of a batch of seedlings would be likely to produce white flowers the following year.

If the leaf stem shows purple, I understood, the flowers will be purple. Plants with pale green leaf stems are more likely to produce white flowers. Going by this, I picked out a number of seedlings several years ago and planted them around the place. A few of them did turn out to have white flowers, but not all. Some produced blooms of a pale, washed-out pink with nothing to recommend it, and some were just the ordinary purple shade.

After this experience, I didn’t bother with systematic seedling selection. I simply let a few plants produce seed, and left a few of the resulting seedlings where they came up, in spots where a foxglove would be an asset to the scene.

Right now, I have two foxgloves with white flowers. The biggest and most impressive one has dark purple leaf stems. So does the smaller one.


Clearly, the idea that leaf stem colour predicts flower colour is dead wrong.


Just for fun, I checked the leaf stems of a rather impressive purple foxglove also in bloom now.


Its leaf stems have only a slight tinge of purple. If I had looked at them before the plant bloomed, I would very likely have predicted pale or maybe even white flowers.

So much for that notion. Consulting that fount of info, Wikipedia, I find this in the article on Digitalis purpurea:

The colours of the petals of the Digitalis purpurea are known to be determined by at least three genes that interact with each other.[6]

The M gene determined the production of a purple pigment, a type of antocianin. The m gene does not produce this pigment. The D gene is an enhances of the M gene, and leads it to produce a big amount of the pigment. The d gene does not enhance the M gene, and only a small amount of pigment is produced. Lastly, the W gene makes the pigment be deposited only in some spots, while the w gene allows the pigment to be spread all over the flower.

This combination leads to four phenotypes:

  • M/_; W/_; _/_ = a white flower with purple spots;
  • m/m; _/_; _/_ = an albino flower with yellow spots;
  • M/_; w/w; d/d = a light purple flower;
  • M/_; w/w; D/_ = a dark purple flower.

If I could understand the above and make practical use of it, I might be able to predict flower colour in foxglove seedlings. I’d probably get more white flowered plants from my self-sown foxgloves if I pulled up any purple types before they reached full bloom, so as to keep their pollen out of circulation. But I am constitutionally incapable of pulling up (i.e., killing) any plant that is vigorously growing and about to bloom. I even have trouble dispatching certain weeds. So I’m not likely to go around “roguing out” any purple flowered plants if they look good.

In the meantime, I’ll enjoy the blooms I have, and let the white one (which, according to the above, is the white with purple spots phenotype) produce seed. Foxgloves seed like crazy. I’ll harvest a small quantity of seed and sprinkle it around the garden, and in 2019 I’ll see the results.

Update: another, rather small, foxglove plant has since bloomed with white flowers. These are pure white, and without spots of any kind. And the leaf stems are very pale green, without a trace of purple. So now I’m wondering if the leaf stem colour is somehow linked to the colour of the spots within the flower. Observing stuff like this makes gardening — even with common plants — interesting.


  1. Audrey, you have me wanting to check each foxglove stem in my garden now! Thanks for the good info and for sharing the research you’ve done on this, although some is greek to me. 😉

    Liked by 1 person

  2. I found this post while doing some research about foxglove – I used to be a teaching assistant for a botany class, so I might be able to help with the genetic info! To break this down:

    M/_; W/_; _/_ = a white flower with purple spots;
    m/m; _/_; _/_ = an albino flower with yellow spots;
    M/_; w/w; d/d = a light purple flower;
    M/_; w/w; D/_ = a dark purple flower.

    “Phenotype” just means appearance. “Genotype” is the combination of alleles a plant has to give it that appearance.

    An allele is a version of a gene. All foxgloves will have two copies of the M or m, W or w, and D or d alleles (one from each parent – although, foxglove can “self”, or breed with itself, so the “two parents” might actually be a single plant!).

    There are two versions of the gene that controls the anthocyanin purple pigment. The M allele allows the plant to produce purple pigment, and the m allele does not. The M allele is dominant, so if a plant has one copy of M, it will be able to produce purple pigment. That means that foxgloves with M/m or M/M can both produce purple pigment, but foxgloves with m/m cannot. The chart above shows this with “M/_”, meaning the plant need one M allele to have that phenotype. m is recessive, like blue eyes in people, so foxgloves need two copies of this allele to show that appearance/phenotype. Foxgloves with two copies of the m allele cannot produce purple pigment, so they are ALWAYS albino, and the offspring of two albino m/m foxgloves will ALWAYS be albino, too.

    There are two versions of the gene that controls the amount of anthocyanin purple pigment deposited. The D allele allows more pigment to be deposited, so flowers with the D allele will be dark purple. Flowers with the d allele will be light purple/pink.

    There are two versions of the gene that controls the location of purple pigment on foxgloves. W allows purple (it it if produced, meaning the plant has the M allele) ONLY on the spots. w allows the purple all over the flower, but it is recessive, so a plant must have two copies to have purple anywhere other than the spots.

    What does this mean if you want white flowers? If you breed albinos to albinos (m/m alleles), you will always get albinos. If you do a dihybrid cross of two purple flowers (M/m, w/w, D/d), the offspring will be in a ratio of 9:3:4 purple:pink:albino flowers. A trihybrid cross of flowers with (M/m, W/w, D/d), which would be white with purple spots (M/m, W/w, d/d, or M/M, W/W, D/D can also be white with purple spots because the dominant M allele means they produce purple pigment and the dominant W allele means they are white except for throat spots), you will get a ratio of 36:9:3:16 white with purple spots:purple:pink:albino flowers.

    The only sure-fire way to get white flowers is to breed albinos to albinos. If you want white flowers with purple spots, breeding those to other white with purple spot flowers will produce mostly white with purple spot offspring, but some will be recessive w/w and thus be purple or pink instead. I have found some very dark-leaved albino plants, so I don’t think leaf color is the best way to determine flower color!

    Liked by 1 person

    1. Thank you for this explanation! I don’t try to influence what my foxgloves are up to, but I let only the white ones (with and without spots) actually ripen seeds, so I now have more of those, with a few purple/pink ones for variety. I suppose if I removed the latter as soon as their colour became visible, I would have only white ones, but I find it impossible to kill a healthy plant with buds. I’m not surprised that you confirm that the notion of leaf colour indicating flower colour is wrong. Thanks again!


Comments are closed.