2 I 1 
Modern Systems of Classification. 
botany as belonging to the realm of philosophical science, and not as 
a complex of pigeon-holes for the docketing of plants, will be disposed 
to resent the too-frequent intrusion of mere “ praktiske Grunde ” 
into a department of science. It seems to us that Engler’s arrange¬ 
ment of the Centrospermae is a very logical one; and the only 
complaint we have to make with regard to it is that we would prefer 
to look upon the group as a complex of orders, and to elevate to 
the rank of orders the four sub-orders which, according to Engler, 
constitute the whole group. There are fewer difficulties in regarding 
the Chenopodiaceae as the most primitive of the Centrospermae than 
in regarding the Caryophyllaceae in that light. There are so many 
points of specialisation in the Caryophyllacese—the straight ovary 
of Diantlius, for example—that it seems reasonable to regard the 
family as a whole as having sprung from one in which those 
specialised characters are absent; and with regard to the Ai'zoaceae, 
including Mesembryantheinum, the whole flower is so extremely 
specialised in all its parts, that no one, we presume, would dream 
of regarding its numerous stamens, carpels and ovules as primitive. 
The position of the Cactaceae has long been a matter of con¬ 
troversy among systematists ; and the family has been placed in 
many different positions by different botanists. Both Wettstein 
and Warming place it immediately after the Ai'zoaceae, though 
Wettstein regards it as coming within the limits of the order 
Centrospermae, whilst Warming places it in a separate, but allied, 
order, the Cactales. Thus both these authorities dffer with 
Engler who places it immediately after his Parietales. There was 
some discussion on the Cactaceae in this Journal a little time 
ago. The discussion involved a consideration of the systematic 
position of the Cactaceae ; and Mrs. Arber 1 stated that she was 
inclined to think it “a justifiable hypothesis” that the Cactaceae 
are derived from the same stock as the Nymphaeaceae. In Wett- 
stein’s arrangement, Mrs. Arber’s view receives no support; for 
although the Cactaceae are here (to quote Mrs. Arber) “ placed 
immediately before the Polycarpicae,” there is no implied con¬ 
nection ; but rather the reverse. The Cactaceae, according to 
Wettstein, are the last and most specialised of his Monochlamy- 
deae ; and the Polycarpicae are the first and least specialized of 
his Dialypetalae. It is not natural to derive the lowest members of 
one group from the most specialised members of another ; and as 
a matter of fact Wettstein connects his Polycarpicae with the 
Hamamelidales which he places in his Monochlamydeae. On the 
the other hand, although the Opuntiales (with the single family 
Cactaceae), are placed apparently high up among Engler’s orders, 
yet a study of his major divisions, which are unfortunately not 
named and thus elude the attention of non-systematists, leaves it 
possible to regard the Parietales and the Opuntiales together as 
having sprung from Ranalian ancestors ; and thus the feature which 
Mrs. Arber ( loc . cit.) points out as being common to the Ranales 
and the Opuntiales are much more easily explicable on Engler’s 
view of the relationships of these families than on Wettstein’s and 
Warming’s. 
Incidentally, it may be worth while emphasising here that there 
are several cases where Engler has been misunderstood by British 
1 New Phytologist, IX, p. 333, 1910 (see also de Fraine, in Ann. 
Bot., XXIV, p. 125, 1910). 
