144 SCULPTURE OF THE TESTA OF GERMAN JUNCACEA. 
transverse ; besides, the seeds of the former species are larger and more 
pointed than they are in the latter; and if the magnifying power be 
still more increased, the fields of the meshes of J. spherocarpus are 
found to be smooth, and in J. Tenageja furnished with extremely small 
elevations, which may be owing either to very minute granules or to a 
system of extremely delicate sunken and netted lines. My attention 
being thus directed to this point, I examined a series of other seeds, 
and soon became convinced that the more delicate structure of the testa 
presented very important characters for diagnostic purposes. On 
learning from Professor Alexander Braun that Dr. Engelmann, of St. 
Louis, Mo., was working up the Junxcacee of North America, I did 
not fail to draw his attention to this character, and was glad to find 
that in his ‘ Revision of the North American Species of Juncus,’ pub- 
lished in the ‘Transactions of the Academy of Science of St. Louis,’ 
1866, vol. ii. n. 2, it had been well attended to, and been raised to 
an important rank. In the following I shall therefore use as my basis 
Engelmann’s paper, which enjoys priority of publication; but I 
thought it necessary to make the foregoing remarks, in order to show 
how it comes to pass that both of us have made, independently of each 
other, the same observations, 
According to the surface of the seeds, Dr. Engelmann distinguishes 
three principal groups, which he defines as follows :— 
1. Semina reticulata; vix s. distincte apiculata. 
2. Semina transverse lineolata, levissime costata; vix s. distincte 
apieulata seu breviter caudita. 
3. Semina costata, plus minus caudata. 
“Coste” he calls the more or less developed longitudinal lines. 
When these alone are prominent, and connected by few and obscure 
transverse lines, the seeds are naturally called ** semina costata.” In 
explanation of the other two terms I add Dr. Engelmann's own words : 
— When the ribs are fewer, are wider apart, and united by transverse 
ridges, so to form somewhat rectangular meshes, I call the seeds 
‘ semina reticulata ;’” and “a large number of Junci exhibit (in fully 
ripe seeds) a very delicate but regular transverse reticulation without 
very distinct ribs—' semina lineolata: ” 
At first sight this classification seems to have much to recommend 
it, for it resolves itself into this :— 
Seeds with longitudinal ribs only (or nearly so)—* semina costata.” 
