214 DR J. M. MACFARLANE ON THE 



' L rwfa ena no ^ "uncommon to find elongated cells over the veins. The lower epidermal cells (fig. 9) 

 -• p ^ ll * g t" a ^'e more sinuous in outline, thinner in their walls, and of smaller size. The stomata 



veitckii. ' 



5. Ph ( j 1 1 i ? 8ia buxi - which lie amongst these are irregularly disposed and freely exposed on the surface. Under 

 Zeiss' D objective with 4 eyepiece, ten to eleven are seen over the field. The upper epidermis 

 of 3 shows cells of varying size, but with straight, yellowish, and thick walls, abundantly 

 traversed by pore canals. The lower epidermis has to the naked eye a whitish waxen 

 appearance, and this is found from microscopic study to be due to extremely minute 

 wart-like papillae that are deepest over the outer area of each convex cell surface. 

 The presence of these breaks up and diffuses the light. They are not wax excretions 

 since they are unaltered by all wax tests ; neither do they appear to be pure 

 cuticle, though their persistence in an unaltered state after many chemical tests suggests 

 a peculiar modification of cuticle. I have at times noticed on surface view what seemed 

 to be the homologue of them in Lapageria, though of extreme fineness, but vertical 

 sections have as yet failed to reveal their undoubted presence. 



It occurred to me that this peculiar formation in PJdlesia might be a development 

 suiting it to climatic surroundings, and that other genera from the same region might agree 

 with it. The first plant selected for comparison was Astelia racemosa, which is very 

 different in its general features though included in the same natural order. Its lower 

 leaf epidermis showed exactly similar surface formations, so that a further examination 

 of plants from the same region is highly desirable. 



The lower epidermal cells of 3 (fig. 7) have strongly convex surfaces, are straight 

 walled, and, as pointed out by De Bary, # the stomata are in rows, while " the slits run 

 perpendicular to the axis " of the organ. They are so deeply sunk, however, that the 

 guard cells are entirely hid, and the stomatic orifice is seen as a faint slit in the 

 depression between adjoining epidermal cells. Commonly one stoma alternates with 

 each epidermal cell, so that thirty-two to thirty-four are visible under Zeiss' D with 

 4 ocular. 



The outline of the upper epidermal cells of 2 in some specimens examined were 

 decidedly more like those of 3 than 1 (Plate III. fig. 5), the waviness of the walls being 

 very slight, and the thickening considerable ; but even in such, one could readily trace 

 the effect of Lapageria parentage if preparations of the three were placed side by side. 

 In other material, however, the outline was as exactly intermediate as if one had 

 attempted carefully to draw it so. This is one of the many examples which one constantly 

 encounters of variability in a hybrid, as in species, and demonstrates the need of exact 

 comparison as to age, position on the stem, food supply, &c, for I believe that consideration 

 of such points probably explains the apparent discrepancy. 



Apart from this, however, it is generally to be noticed that both upper and lower 

 epidermal cells of the hybrid are equal to, if not larger than, the largest of either parent. 

 Those of 1 are on the average larger than those of #, but in the hybrid they may be larger 

 than in either parent. Now, from Kolreuter's time onward, the increased luxuriance of 



* Cornp. Anat. Phan, and Ferns, Eng. ed., 1884, p. 45. 



