322 PAEKS AND PLEASUEE-GEOUXDS. 



synthesis of characters, are made to comprise alliances 

 and classes. These successive combinations, then, give 

 rise to what we may call generic^ cardinal^ allied^ and 

 classiJiG groupings ; the last, however, we mention 

 chiefly for the sake of completeness, because, though 

 carefully preserved, it will be almost invisible, except 

 in one or two cases, on account of the multitude of 

 subordinate members which it includes. Every one 

 will ad^nit the propriety of planting each genus by 

 itseif, and this accordingly is the first step in the ar- 

 rangement. But the genera, as spread out on the 

 ground, may be combined into orders ; and these 

 orders, again, may be made to occupy such positions 

 that the allied races, or "alliances," may be marshaled 

 together in space, and may be exhibited to the eye in 

 living presence as well as presented to the understand- 

 ing in theory. It is to the proper collocation of the 

 orders and alliances that attention must be chiefly 

 directed in the scientific treatment of the arboretum. 



A consideration of the synopsis will bring out other 

 important results. By tracing the numbering of the 

 orders and alliances, as quoted from "The Vegetable 

 Kingdom" of Dr. Lindley, the reader will observe 

 that many of the orders, silliances, and even classes, 

 which exist in a comjDlete system of j^lants, are ex- 

 •cluded from our list by the conditions imposed by our 

 cold climate and the arborescent growth of the subjects 

 employed. The orders that remain are often found 

 in a sadly mutilated state. In the Fabacew, for exam- 

 ple — the old "pea tribe," or the Papilionaceous order, 

 as it was wont to be called — the genera, as estimated 

 in " The Vegetable Kingdom," amount to 467, while 

 we have been able to assign only twenty-five as 



