as having asingle longitudinal ‘‘slit’’. Misinterpretation 
of the material led him to this conclusion. It will be ob- 
served in the accompanying figures that Codonotheca 
spores have two furrows as have all other members of 
the Whittleseyineae. Halle remarks (p. 11) that these 
furrows ‘‘might be supposed to mark the junction be- 
tween the part of the wall facing outwards in the tetrad 
and the two radial walls.’ 
Codonotheca caduca Sellards seems to be restricted to 
the environs of Mazon Creek. It has been found at Mazon 
Creek, Braidwood, Wilmington and Morris. Somewhat 
similar, but not sufficiently preserved, specimens have 
been collected in the Upper Allegheny formation of Ohio 
and western Pennsylvania. Turner (76) has figured pol- 
len grains found in the flame-etched surface of polished 
anthracite coal from the Ross coal of Nanticoke, Penn- 
sylvania. The Ross coal is of Middle Allegheny age. It 
is impossible to identify this pollen specifically with that 
of Codonotheca, but there is no doubt that the pollen be- 
longs to one of these medullosan fructifications. 
Recently Chamberlain (4) has described Neuwropteris 
decipiens as bearing Trigonocarpus seeds. He was led to 
this opinion by the celebrated reconstruction in the Field 
Museum of Natural History in Chicago. Professor Noe 
informs me that unfortunately the seeds have not been 
found in attachment. It is probably true that Newropteris 
decipiens is a pteridosperm, that it bore polleniferous 
structures of the Codonotheca type and that it had seeds of 
a Neuropterocarpus of Rhabdocarpus shape and structure. 
Crossotheca, unlike Codonotheca, is widely distributed 
and has figured prominently in phylogenetic speculations 
on the seed-torms. In addition, there are a number of 
points involved which stimulate reasonable doubt regard- 
ing the pteridospermous nature of at least a few species 
of Crossotheca. 
[162 ] 
