HALLE, ON THE SPORANGIA OF SOME MESOZOIC FERNS. 9 



Returning to the Scanian specimen of D. fecunda, we may 

 suppose that the venation of the fertile pinna was somewhat 

 similar to the one just described in the case of D. marantacea, 

 only that in the former the thicker veins are alone preserved. 

 At any rate it is certain that the sori are more often placed 

 between the veins than over them, and that the relation be- 

 tween the sori and the veins is not so regular as in the recent 

 Marattiaceae. 



It might perhaps be suggested that the peculiar relation 

 of sori and veins may be illusory and that the supposed veins 

 between the sori are really the marks of a raised wall-shaped 

 indusium similar to that of Danaea. A close examination of 

 the veins, however, shows that this is not the case, the veins 

 branching and joining the rachis in a manner inconsistent with 

 the idea of an indusium. 



The sporangia can easily be detached from the rock. 

 The sporangial wall shows very little resistence to chemical 

 agents. If the sporangia are treated in the usual way with 

 Schultze's mixture and afterwards with ammonia, their walls 

 are dissolved and the spores set free. 



The spores (pi. 1, fig. 11) are of the tetrahedral type, 

 globular, with the usual triradiate marking. Their surface is 

 smooth without any papillae or other sculpture. The dia- 

 meter is usually 0,o6 — 0,07 mm, but many spores are consider- 

 ably smaller. 



As the spores are in most cases easily separated from each 

 other, I have tried to count them in order to settle the spore- 

 output per sporangium. Three apparently intact sporangia 

 were selected, and the actual result of the counting of the spores 

 was 1 100, 1 175, 1 159. Of the typical numbers of spores, 

 1 024 comes nearest to the result obtained. It is possible that 

 the result of the countings may be too high: the sporangia 

 often overlap, and in attempting to remove the entire sj)or- 

 angia, portions of adjacent ones may have been left adhering 

 to them. At any rate the countings give an approximate idea 

 of the spore-output. It is not probable that the number of 

 spores per sjjorangium can be as high as the next higher typ- 

 ical number or 2 048. In respect to the size of the sporangia 

 the spore-output is certainly surprisingly small. The num- 

 ber of spores in Angiopteris is given by Bower (1908, p. 

 642) as 1 450, and yet the sporangia of Danaeopsis, as far as 



