BOOK REVIEWS. 



131 



caution to the verge of obscurantism. Those trained in Ray's attempt 

 at a natural arrangement surrendered most unwillingly to the 

 revolutionary artificial grouping of Linnaeus ; and half a century 

 later the followers of the Swedish botanists were even more determined 

 in their resistance to the Natural System of the Jussieus and De 

 Candolle. Now, after another half-century, we are showing almost 

 equal reluctance to the adoption of changes " made in Germany." 

 It has long been admitted that the division of Gymnosperms from 

 Angiosperms is far more fundamental than that between Mono- 

 cotyledons and Dicotyledons ; and yet we still too often see the 

 Coniferae sandwiched between Amentaceae and Typhaceae. It has also 

 long been accepted as a pious opinion that the Incompletae are a 

 heterogeneous assemblage which ought to be distributed to the neigh- 

 bourhoods of groups of chlamydeous plants to which they show 

 affinities in some cases obvious. Here, however, comes the difficulty. 

 We are not prepared to acknowledge that Dr. Engler, or anyone else, 

 has solved all the problems as to the affinities of all the Families ; 

 and some are, therefore, unwilling to adopt his system, in spite of its 

 close approximation to the desired accurate representation of these 

 affinities. The scheme was first put forward in 1892, a year before 

 the completion of that of Bentham and Hooker ; but it has undergone 

 many improvements since then, down to the seventh (1912) edition 

 of the " Syllabus der Pflanzenfamilien." In 1897 Dr. J. C. Willis gave 

 a full synopsis of the system, with the characters of the Orders (then 

 called Cohorts), in his " Manual of Flowering Plants," and ten years 

 later Messrs. Britten and Rendle thought it well to append to their 

 "List of British Seed-Plants and Ferns," though it is arranged on the 

 old system, a " Sequence of Orders (Families) according to recent 

 views of affinity," which is in the main Engler's. Lastly, only a few 

 months back, the Cambridge University Press issued an absurdly high- 

 priced little book — four shillings for 139 pages — by Dr. Humphrey 

 G. Carter — giving not only the Families, but also the characters of 

 the genera of British plants arranged on Engler's system. This has 

 well-nigh rendered Mr. Grove's somewhat cheaper book — one shilling 

 for 55 pages — unnecessary ; but the latter is a convenient pocket-book 

 and includes the names of the more important foreign genera. With 

 Engler's sequence of Families we have only one quarrel. We certainly 

 prefer our English separation of Fumariaceae and Papaveraceae ; and, 

 though we accept the position of Droseraceae between Resedaceae 

 and Crassulaceae (among British Families), we consider the lumping 

 of Sarraceniaceae, Nepenthaceae, and Droseraceae into one Order 

 Sarraceniales to be an example of the too great reliance that German 

 botanists have always placed upon purely physiological analogies. 

 We do not like the separation of Balsaminaceae from the Geraniales ; 

 but most of all do we object to the placing of Cucurbit ales between 

 Dipsacaceae and Campanulaceae, in lieu of its former position, with 

 Passifloraceae and Begoniaceae, in the neighbourhood of Parietales 

 and Myrtiflorae. 



K 2 



