eusporangiate sorus with a prodigious production of very 
small ellipsoidal spores (0.012 mm. 0.015 mm. ). Those 
species bearing Crossotheca fructifications have large 
spherical spores. Recently Halle (6) has found a species 
which he named Pecopteris wongu trom the Permo-Car- 
boniferous of Shansi Province of China. This species, 
clearly referable to the miltoni-abbreviata group, may 
prove to be seed-bearing. At least one specimen seems 
to have a single smooth rhabdocarp-like seed attached 
laterally. Halle is unwilling to accept this evidence with- 
out reservation. 
The real reason why Crossotheca is held to be phylo- 
genetically important is because it is believed to be the 
pollen-bearing structure of certain pteridosperms. To be 
specific, the microsporangiate fructification of the classic 
Lyginopteris oldhamia was believed by Kidston (9) to 
be a Crossotheca. Miss Benson (2) at once challenged this 
contention by showing how Telangium could as well be 
the microsporangiate structure of Lyginopteris. 
Actually, Miss Benson has called attention to the 
weakest point in the argument, because the structure 
found by Kidston, though a true Crossotheca, does not 
belong to Lyginopteris oldhamia. The species has been 
renamed Crossotheca kidstont by Crookall, (5) but it 
probably belongs to another lyginopterid. 
It is interesting to note that Te/angium shows a deti- 
nite resemblance in structure to certain members of the 
W hittleseyineae. However, the evidence indicates that 
the synangium is bilaterally symmetrical. 
Crookall (5) has shown that Crossotheca fructifica- 
tions are borne on foliage of both the Pecopteris and 
Sphenopteris types, but he cites an opinion of Mr. Hem- 
ingway that in every case where proof of attachment can 
be obtained, this attachment has been pecopterid. 
Hirmer (8) has arranged what he considers to be a 
[ 167 ] 
