314 BOTANICAL GAZETTE [OCTOBER 
selection has resulted in a race of plants degenerate from the standpoint of their 
power to resist unfavorable conditions. In the fields, even under usual condi- 
tions, a large percentage of the plants wilt on hot days. The reduction of thick- 
ness of the leaf is held in large measure responsible for the occurrence of the 
mosaic disease. A lesser though important influence is also attributed to the direct 
action of the soils —H. HAsSsELBRING. 
Microsporangia of Pteridosperms.—In 1905 Krpstron published a prelimi- 
nary note on the occurrence of microsporangia in connection with the foliage 
of Lyginodendron.? He has now published the full paper,® with detailed dis- 
cussion and illustration. He first elaborates the evidence that Sphenopteris 
éninghausi Brongn. and Lyginodendron Oldhamium Williamson are identical 
plants, and of course Crossotheca Héninghausi is the fertile pinnule of the former. 
It follows that the sporangia found on this species of Crossotheca are the micro- 
sporangia of Lyginodendron Oldhamium, a well-known pteridosperm. This rules 
out Miss Benson’s claim that Telangium Scotti represents the microsporangia 
of this Lyginodendron. 
The microsporangia are borne on modified pinnae, associated with sterile 
pinnae. The fertile pinna is oval, entire, on a short thick pedicel, and on the 
under side of the blade six to eight fusiform bilocular sporangia are borne, which 
bend inward at an early stage so that their pointed apices meet, forming a asi 
of hemispherical sorus. Later they spread apart and appear as a fringe hanging 
from the margin of the pinnule. In all cases the microspores are present, but 
internal structure seems to be evident. The genus Crossotheca contains eight 
species, a new one being described in this paper, and perhaps it is safe to assume 
that all of them belong to the pteridosperms. 
In a general discussion of the occurrence of fern-like plants, Krpsto 
to the conclusion that the pteridosperms (including all Cycadofilices) are 
undoubtedly the oldest group of fern-like plants of which we have evidence, 
being plainly represented in the upper Devonian; that in the Lower Carbonu- 
erous pteridosperms were still dominant, true ferns being feebly represented, 
if at all, by the Botryopterideae; that in the Upper Carboniferous the same 
relative representation continued. It seems highly improbable, therefore, as 
the pteridosperms could have descended from true ferns, and KrpsTon is inclined 
N comes 
to believe that there is no more relationship between the two groups than that 
of a common ancestry for pteridosperms and eusporangiate ferns. In er 
ferns a 
up his conception of the most probable lines of descent of the existing ‘ 
cycads, he indicates three lines: an independent one, leading from Botryopterideae 
(of unknown origin) to the leptosporangiate ferns, and two lines converging 12 
common but unknown ancestral forms, one leading to the Marattiaceae, the other 
through pteridosperms to the cycads.—J. M. C. 
7 Bor. GAZETTE 41:219. 1 ‘ arks 
8 Kipston, Rosert, On the microsporangia of the Pteridospermae, with ar ve 13- 
on their relationships to existing groups. Phil. Trans. Roy. Soc. London B. 198:4 
445. pls. 25-28. 1906. 
