204 DR J. M. MACFARLANE ON THE 



the occurrence, artificial production, relative fertility, variability, and external appear- 

 ance of hybrids. The special aim in this paper will be to compare their tissues and cell- 

 elements minutely. Short synopses of my earlier results were given in the Gardeners' 

 Chronicle for April and July 1890. Until after the publication of these I was not aware 

 that some advance had already been made in the direction indicated, and my best thanks 

 are due to Dr Masters for calling my attention to one or two publications on the subject. 



In 1831 Professor J. S. Henslow compared* a hybrid Digitalis with its parents in 

 a wonderfully minute way, when we consider the degree to which histology had advanced 

 in his time. He demonstrated that in the size and shape of the hairs and other structures, 

 the hybrid was intermediate between its parents. Wichura t and Kerner J have proved 

 that the same is true of Willows and Pulmonarias respectively. 



But to Wettstein § belongs the credit of having compared the leaves of four coniferous 

 hybrids with those of the parents in general tissue arrangement. His descriptions and 

 illustrations are all that could be desired, and had he carried out the comparison more 

 minutely, much that is included in the present paper would have been superfluous. He 

 showed from transverse sections of the leaf that each hybrid is exactly intermediate between 

 its parents in the number of stomata exposed on section, the depth of the epidermal cells, 

 and the number and arrangement of the sclerenchyma elements of the bundles. 



Since the publication of the preliminary account of my results in the Gardeners' 

 Chronicle, a series of communications from Monsieur Marzel Branza has appeared || which 

 deal, like those previously referred to, with the tissue masses only. I have not had access 

 to any of the seed hybrids he describes, but one plant, Cytisus Adami, which we have both 

 been able to examine, is either wrongly described by him, or its tissue and cell arrange- 

 ments differ remarkably in the examples that we have each obtained. As my results 

 have been drawn from detailed study of the parts of thirteen specimens, which agree 

 exactly with each other, I am compelled to accept the former explanation. 



While carrying out a minute comparison of upwards of sixty hybrids with their 

 parents, I have been led to adopt certain precautionary measures which must be kept 

 constantly in view if one is to arrive at safe results. These are as follows : — 



(a) Average Organismal Development and Deviations from it. — It is now recognised 

 by botanists that every species exhibits a sum- total of naked- eye characters which dis- 

 tinguish it with greater or less precision from allied species. These are duly given in every 

 local Flora. But further, specific features — alike macroscopic and microscopic — which are 

 of great importance, are passed over. Radlkofer IT has already insisted that the anatomi- 

 cal method must be applied to the study of species, and I have pointed out that this is 

 equally true of sub-species or varieties.** But it is the sum-total or accumulation of 

 minute peculiarities which gives specific identity to any organism, and it is to be expected 

 that evident or naked-eye variations will often have their commencement in trivial struc- 



* Camh. Phil. Trans., vol. iv., 1833. t Bastardbefruchtung, 1865. 



I Monographia Pulmonar., 1878. § Sitz. der Kaiser. Alcad. der Wissen., vol. xcvi., 1888. 



|| Comptes Rendus, tome cxi. No. 6, 1890 ; Revue Gdnfrale de Botanique, tome i. Nos. 19, 22, 23. 



IT Akad. de- Wissenschaften, Munich, 1833. ** Trans. Bot. Soc. Edin., vol. xix., 1891. 



