SPORE-ENUMERATIONS 643 



sporangium was uniformally large. Passing from these to the prevalent 

 Leptosporangiate Ferns of the present day, the output falls with some 

 degree of consistency, and the striking fact emerges that, so far as 

 observation goes, no Polypodiaceous Fern has a larger output than 64, 

 while in many cases it is smaller. The evidence points to a general 

 reduction in the course of Descent. 



But variations occur within near circles of affinity, and in accordance 

 with the above generalisation these have a special interest in those 

 families which hold an intermediate position. This is seen within 

 moderate limits in the Gleicheniaceae, Osmundaceae, and Schizaeaceae, 

 in which the numbers approximate for the most part to the typical 

 numbers 512, 256, 128. In some cases it is difficult to see any circum- 

 stances which serve as an explanation ; thus the variation between 

 Gleichenia flabellata (512-1024) and Gl. dichotoma (256 or more), is not 

 susceptible of ready explanation, unless it be connected with the more 

 numerous sporangia in the sorus of the latter. Nor is that within the 

 genus Lygodium. In other cases, however, the conditions of life and 

 the structure of the assimilating apparatus throw some light on the 

 matter ; thus Todea barbara, with its thick assimilating leaves, gives 

 higher numbers per sporangium than T. superba and hymenophylloides^ 

 with their thinner pellucid leaves. But the most interesting series in 

 this connection are the Hymenophyllaceae, for the limits of variation in 

 number are here very wide, ranging between such numbers as 421 and 

 32. Here the variation in number of spores per sporangium runs fairly 

 parallel with the size of the sporangia, the smaller number of spores 

 being contained in smaller sporangia (compare Tr. reniforme and 

 Tr. pinnatum'). Further, there is a correlative elongation of the recep- 

 tacle, on which there is thus space for a larger number of the smaller 

 sporangia to be inserted and produced in succession. On these characters 

 the Hymenophyllaceae may be laid out as a series, extending from types 

 chiefly of the less specialised genus Hymenophyllum with short receptacle, 

 large sporangia, and large output of spores from each, to types of the 

 more specialised genus Trichomanes, with longer receptacle, smaller spor- 

 angia, and smaller output of spores from each. The former link on 

 naturally to the Gleicheniaceae in the characters named, as well as in 

 the general appearance of the sporangia : it seems not improbable that 

 in the Hymenophyllaceae we may see a series of specialisation in which 

 the " filmy " character is the most salient vegetative feature, and that 

 this has carried with it, as in the genus Todea, a decrease in size of 

 the sporangia, and in the number of the spores produced from each. 



Taking the facts thus acquired from the Simplices and (Iradatae they 

 show that within circles of near affinity there may be a wide margin of 

 variation in output of spores per sporangium, such as a theory of phyletic 

 reduction of the sporangium would demand : and this goes parallel with 

 the increase in number of the sporangia in the sorus, a decrease in size 



