73 



Cucurbitaceae are placed not far from Passifloraceae, as by 

 Linnaeus, Jussieu, Bentham and Hooker, etc. 



Some of Engler's orders, such as Geraniales, Sapindales, 

 Parietales, Contortae and Tubiflorae are divided by both Rendle 

 and Hutchinson. Along this line Moss wrote in 1912 "Warm- 

 ing's orders and families are smaller than those recognized by 

 Engler. The immense size of some of Engler's orders, e. g. 

 his Geraniales, make them almost if not quite unworkable." 

 With the exception of Cucurbitaceae, the Sympetalae are main- 

 tained as a group by both authors, though both admit that 

 these are related along different lines with polypetalous groups. 



In Rendle's book some of the smaller families have been 

 omitted, and more space given to families represented in Britain. 

 Hutchinson, who includes all families, favors going "slightly 

 further even than Engler" in dividing them; "the delimitation 

 of families, of genera and species is sometimes very much of a 

 matter of taste and personal idiosyncrasy; also of judgment 

 and experience." Bentham and Hooker had 160 families of 

 Dicotyledons. In the Pflanzenfamilies there are about 250 

 families of this group, 255 in the 9-ioth edition of the Syllabus 

 (1924); ten of these are not taken up by Hutchinson, namely: 

 Hippocastanaceae and Bretschneideraceae (H. in Sapindaceae), 

 Roridulaceae (in Byblidaceae), Pentaphylacaceae (in Theaceae), 

 Elaeocarpaceae (in Tiliaceae), Hydrocaryaceae (in Onagraceae), 

 Desfontaineaceae (in Loganiaceae), Brunoniaceae (in Goodeni- 

 aceae), Strasburgiaceae, Hoplestigmataceae. 



Hutchinson has 264 families; the 19 additional ones are: Win- 

 teraceae, Schizandraceae, Cabombaceae, Circaeasteraceae, Sar- 

 gentadoxaceae, Molluginaceae, Illecebraceae, Samydaceae, Saur- 

 auiaceae, Greyiaceae, Grossulariaceae, Hydrangeaceae, Caesal- 

 piniaceae, Mimosaceae, Barbeyaceae, Corylaceae, Cannabin- 

 aceae, Monotropaceae, Selaginaceae. 



In the outline of orders and families reasons are given for the 

 arrangement adopted. Lines separate groups of orders, the 

 larger groups are not named; this was the suggestive method 

 used by Bernard de Jussieu. As appears from the diagram 

 an important feature of Hutchinson's system is the division of 

 Archichlamydeae into a woody and herbaceous series. Thus it 

 "becomes possible to establish more natural groups of closely 

 allied families." In this connection importance is attached to 



