origin. ‘he only described leaf which seems to be char- 
acterized by similar features is Cordaites crassus Grand’ 
Eury from Grand Croix in central France. 
There is some uncertainty, however, concerning the 
validity of the numerous species based on leaves which 
have been provisionally recognized. Numerous of the 
lowa coal balls contain leaves that appear to be much 
nearer to Cordaites felicis Benson (2) which is most prob- 
ably the foliage of Mesorylon multirame. To Benson’s 
species belong at least three and perhaps four nominal 
species. 
Felix in 1886 (6) described three types of leaves from 
the Westphalian rocks of the Ruhrgebiet under the 
names Cordaites wedekindi, C.loculosus, and C.robustus. 
He had only a limited number of thin sections at his dis- 
posal and was obliged to name three distinct forms. In 
1912 in an important communication, the late Dr. Mar- 
garet Benson described Cordaites felicis. She was able to 
determine much of the histology of the whole leaf. The 
basal portion of C_felicis possessed a construction similar 
to that of C.loculosus and C.robustus. The upper portion 
resembled considerably C. wedekindi. Benson was not 
able to demonstrate the specific identity of all of these 
forms, but Koopmans (18) has not hesitated, on the basis 
of Dutch material and comparative studies, to unite them 
into C. felicis. There is a point of nomenclatorial priority 
here which must be determined in keeping with standard 
procedure. Koopmans also believes that C. weristeri 
Leclereq (14) belongs to C. felicis. 
Inasmuch as several sections from a single leaf have 
a strikingly variable structure, related of course to the 
portion from which the section was made, it seems un- 
wise to propose a specific name for the figured specimen 
without qualification. 
[13 ] 
