256 Inheritance in the Groundsel 



Flore de France, 1903, recognises no hairy forms. Until colonies of 

 lanuginosus had been grown in my own garden, I had assumed, like 

 most other botanists, that hairiness in groundsels was the result of a 

 more or less direct response to changes in the external conditions, be- 

 longed, in short, to the category of modifications. Lanuginosus, however, 

 is always very, hairy, whatever the external conditions may be. It is 

 almost certainly impossible to raise a colony of these plants in the 

 glabrous condition. Most of the other types are normally so glabrous 

 that hair can, as a rule, only be found by looking carefully for it. The 

 most glabrous forms, however, produce a little hair in the leaf axils and 

 the buds are consequently often slightly hairy. Such hair is, however, 

 soon lost, and the plants appear to be, as indeed they practically are, 

 quite glabrous. In this sense praecow, genevensis, erectus, multicaulis, 

 latifoUus and erectus, radiatus are glabrous. 



Intermediate between these two types are two other forms of 

 hairiness. In these hair is constantly to be found on most parts 

 of the plant until full maturity is reached, when it generally becomes 

 greatly reduced in amount. The hair, even when abundant, however, 

 is never dense enough to mask the stem-colour, and appears white at a 

 distance, as in lanuginosus, yet it is generally recognisable with certainty, 

 except perhaps in very old plants. These forms of hairiness occur in 

 several incompletely investigated types that have been cultivated for 

 several years in the garden. We thus have to recognise four standards 

 of hairiness : — (1) very hairy {H^), as in lanuginosus ; (2) distinctly hairy 

 (if^), as in three types from Burry Green, Horton, and Cross Common 

 respectively ; (3) slightly hairy {H^), as in two types from Cardiff and 

 St Bride's respectively, and (4) the glabrous types {H^). (PI. XVIII, 

 figs. 21, 22, 23 and 24.) 



Grown under constant conditions, side by side in the same garden, 

 the standard of hairiness is maintained for generations. It naturally 

 varies a little in response to seasonal changes, but as all the plants are 

 subject to the same change, the same relative differences appear year 

 after year. 



It may perhaps be advisable to put on record the extent of the 

 experiments which have led to these generalizations. 



The lanuginosus type of hairiness is very distinct, and so is the 

 glabrous condition of such types as praecox and erectus. The inter- 

 mediate types — H"^ and H^ — are more variable. They are sufficiently 

 definite to admit of classification when pure colonies are examined. 

 When segregation of hair factors is taking place, however, it is almost 



