W. F. R. Weldon 49 



So miicli l'or what is known alxiiit ihu previuus history of the raccs used by 

 Mr Batcson aiul Miss Saundeis. It is sufficient to show that even the use of 

 Mendcr.s categories is not enoiigli to bring the phenomena described into any- 

 thing like e.ract accord with Mendel's laws. Betöre going further, wo will exainine 

 a little more closely the conditious included uader the two categories "glabrous" 

 and "hairy." 



By glabrous Professor de Vries understands (at least when speaking of L. 

 vespertina) the abseiice of perfect hairs from all parts of the plant, although he 

 calls a variety " glabrous " when especially in the young state it has " traccs of 

 hairs " here and there. Mr Bateson and Miss Saunders leave ihrlr nieaning a little 

 doubtt'til. ()n p. 16 tliey say "in the glabrous varieties no hairs wore observed on 

 any part of the plants at any time"; but since when speaking of cross-bred forins 

 they always rofer to hairiness as a character of leaf-surface, it is not quite clear to 

 nie whether a cross-bred plant with glal)i-ons leaves and a hairy stem was ever 

 observed by them, and if so, whether they would classify it as glabrous or as hairy. 

 The only glabrous plant I have myself foiind wild {L. vespertiiia) had uo hairs on 

 any part of its surface ; and I am not aware that plants with hairy stems or calices 

 and glabrous leaves have been described. For this reasou the repeated statement 

 that hairiness is a character of leaf-surface is difficult to understand. 



The category " glahrous" whether merely applied to leaf-surface or to the whole 

 plant, certainly iucludes only a small ränge of conditions. It is far otherwise with 

 the category " hai7'i/." In accordance with the differences in habit of the two 

 species, the hairs of L. diurna and of L. vespertiiia differ considerably in their 

 distribution, even when leaf-surface only is considered ; and while the average 

 condition of the hairs in the two "species" is ilifferent, each "species" varies from 

 race to race, and from individual to individual. 



The hairs are multicellular, and niay vary in length, on the same leaf, from 

 aboiit 0'2 nun. to about 19 mm.; the cell sap may be red, or hlue, or nearly 

 colourless ; some of the hairs on a leaf may be glandulär, the percentage of such 

 hairs varjung greatly. Further, when a race of L. diurna and one of L. vespertina, 

 both being hairy, are crossed, the liybrids are said by Gagnepain to be iutermediate 

 betweeu the parent races in the condition of their hairs {Bull. Soc. But. de France, 

 T. XLiv. p. 44-5). 



In Order to avoid the effects of differentiation among the leaves, it is necessary 

 to compare corresponding leaves of the different plants examined. To illustrate 

 the Variation in the frequency with whieh hairs occur in Unit area of leaf-surface 

 Table I. has been compiled. It shovvs the number of hairs per square centimetre 

 of epidermis on tlie lower side of a leaf from the node below that which bears the 

 terminal flower. The observations were made (a) upon a race found in a little 

 copse on the Berkshire (right) bank of the Thames, about four miles by road below 

 Oxford, and close to SandforJ Lock; (h) upon a race found on the slopes of Cooper's 

 Hill, Surrey, also on the right bank of the Thames, but more than 35 miles distant 



Biometrika ii 7 



