EI TATE ERE TNRR 
FLORA AMERICJE BOREALI-ORIENTALIS. 855 
tended to in the American work, but occasionally (although not fre- 
quently) lost sight of in the German one, and to which we would call 
attention, on account of its importance in illustrating the details of the 
fruit and seed : this is the vertical position of the analyses on the plate, 
ə Which should always be that which they have ^ nature, or rather 
according to a generally admitted conventional nature. Whether 
flower itself be erect or drooping (a character meiden indicated in 
the general figure), the point of attachment of each primary organ (that 
is, of each organ proceeding from the main axis of the flower) must be 
considered as its base, and the organ, when isolated, represented as 
erect from that base. In the seed, on the contrary, where the relative 
position of the embryo, radicle, &c., to the fruit is as essential as that 
which it bears to the point of attachment of the seed, all the parts, 
whether isolated or not, should be represented in the position they 
would occupy supposing the fruit to be erect. In those cases where all 
the seeds of a fruit are naturally horizontal, and it is necessary to re- 
present them seen from above, it is always best to draw them with the 
point of attachment on one side, and neither erect nor inverted. So in 
the diagrams or plans of arrangement, the posterior side next the axis 
of inflorescence should be uppermost, and the anterior, next the sub- 
tending bract, below ; or if the flower be terminal, having, consequently, 
no anterior and posterior side, the two last bracteole should be right 
and left. 
In the text, the parallel between the two works ceases. In the 
German, it is a detailed description of the genus, with a few occasional 
observations ; whilst the American text gives us a comprehensive Genera 
Plantarum of the United States Flora, conceived according to a plan 
which appears to us admirably suited to all general works of the 
kind, and executed with the ability and care which its distinguished 
author has shown in all his publications. 
According to the systematic plan alluded to in our former notice, a 
short Latin character is appended of each natural order, followed by a 
ees restricted synonymy and referencés, a general view in 
English of the affinities, geographical distribution, and properties, 
adding, in the case of the more extensive families, a more detailed de- 
velopment of the more important characters, with discussions on mor- 
phological and other points, to which we would specially call attention, 
on account of the critical acumen which they display, and the impor- 
: 222 
