RECENT WORK, ETC. 479 



dons with primary tissue-formation only, to those with a 

 definite zone of secondary thickening. To the latter 

 category the shrubby Irideae must now be added (27). 



Before concluding this article some mention must be 

 made of Schenk's important work on the anatomy of 

 Lianes (28). This is a very full account based on original 

 observation, and abundantly illustrated, of the highly 

 modified structure of climbing plants. The author finds 

 that habit and anatomy do not necessarily go together ; 

 thus twiners and tendril-climbers, so different in their ex- 

 ternal characters, have their most characteristic anatomical 

 peculiarities in common. 



The author is not contented with merely describing the 

 anomalous structures, but also endeavours to trace their 

 origin. In each case there are three possibilities to be 

 considered : the character may be inherited, or may be the 

 direct result of the action of external conditions, or may be 

 adaptive. That the direct action of the conditions of 

 growth has a considerable influence is shown by the fact 

 that cultivated specimens of anomalous climbing plants, 

 grown in the absence of supports, may develop an almost 

 normal structure. Hegler's remarkable experiments on 

 the development of sclerenchyma in consequence of in- 

 creased tension, throw some light on this subject. There is, 

 however, no evidence that such directly acquired characters 

 can be inherited, though Schenk seems to assume that this 

 may happen. 



The author shows that the most modified members of 

 very different families may present quite similar adaptations. 

 Illustrations of this are afforded by the comparison of 

 Baukinia with climbing Malpighiaceae, of Bignouia with 

 Phytocrene, or of Wistaria with the anomalous Menisperm- 

 aceae. Few fields of investigation offer more striking ex- 

 amples of homoplastic characters. 



In those who believe that all characters whatsoever are, 

 or have been adaptive, the distinction between inherited 

 and adaptive characters cannot be an absolute one. The 

 distinction is between recently acquired adaptations, and 

 those which arose in the more remote past, and have come 



33 



