644 GENERAL COMPARISON OF THE FILICALES 
of the sporangium, and in the thickness of its stalk and of the sporangial 
wall, and a progressive change from the early segmentation characteristic 
of the Eusporangiate type to that of the Leptosporangiate. 
As a consequence of such phyletic changes in the constitution of the 
sorus, it is noteworthy how often the output of spores per sorus is 
similar in Ferns which are systematically remote from one another: for 
instance, Marattia fraxinea (45,000) and Polypfodium aureum (57,600) ; 
Angiopteris evecta (14,500) and Hymenophyllum dilatatum (11,500); 
Alsophila excelsa (3,200) and Gleichenia flabellata (3,000). These 
examples show how a similar result may be obtained by various means, 
a large number of small sporangia balancing a smaller number of large 
ones. The similarity of output in such cases may be merely a conse- 
quence of similarity in the powers of the underlying nutritive mechanism. 
The real interest, however, arises when in nearer circles of affinity, with 
varying size of sporangia, and of output per sporangium, the result per 
sorus is kept approximately constant by converse variation of the two 
factors, This is illustrated in the genus Glechenia, and in a less 
precise way in Adsophila excelsa and Cyathea dealbata. But the best 
demonstration of it is seen in the Hymenophyllaceae, undoubtedly a 
very natural series, in which the sorus has a uniform type of con- 
struction, though the size and number of the sporangia, and the length 
of the receptacle are variable. In illustration of this, estimates have been 
made with such accuracy as possible, with the results which are given 
in the subjoined table: 
i S 
Name. sporaneia PEE oraaplamn Output per sorus. 
Hym. Tunbridgense, ®. 20 420 8,400 
Trichomanes reniforme, - 40 256 10,240 
Hym. dilatatum, go 128 11,500 
Trich. radicans, 140 64 8,960 
It thus appears that notwithstanding the great variations of sporangial 
output, the result per sorus is approximately uniform for the cases 
quoted from that very natural family of Ferns. This suggests a true 
biological progression, and it probably does not stand alone, but 
illustrates a principle which has been of wide application in the Fern 
phylum. 
The production of numerous spores is a drain upon the resources of 
the plant. That drain may be relieved within the sorus by the development 
of a succession of sporangia, the demand being thus spread over an extended 
period. In the Simplices the sporangia of a single sorus arise simultane- 
ously: the physiological drain thus comes at one time: this method, which 
is, physiologically speaking, a simple and probably a primitive one, is 
