574 Palaeontologie. 



to enter the pinnule from the pedicel some little distance from the 

 raargin, where it divides at an acute angle into two main branches. 

 Each fertile pinnule usually bore six, rarely seven, bilocular micro- 

 sporangia. They are fusiform and end in a sharp point. The upper 

 portion of the microsporangium was attached to the under surfaces 

 of the limb and probably seated on a lateral veinlet. The sporangia 

 do not seem to have been united among themselves, and though 

 their upper extremities approach each other, they do not appear to 

 have been attached to a common point. The centre around which 

 they radiate is always a solid piece of the matrix. They are about 

 3 mm. long, and about 1.5 mm. wide in their broadest part. When 

 immature the free portion of the microsporangia is bent inwards, 

 but at maturity they spread out, and the sorus has then much the 

 appearance of an epaulet. In some of the specimens, infiltrated with 

 carbonate of lime, the narrow wall dividing the two loculi is clearly 

 seen, and the germinate nature of the loculi within the microsporangia 

 is placed beyond doubt. The cleft by which dehiscence took place 

 passed longitudinally down the centre of the inner face of the 

 microsporangium, and thus the microspores of both loculi would be 

 liberated simultaneously. The microspores are circular or oval, 

 and measure between 50,u— 70;« in diameter. They are very much 

 contracted and folded or crumpled as a rule, and their outer surface 

 is roughened by numerous closely-placed, very minute, blunt points. 

 A distinct triradiate ridge is precent. 



The microsporangia of Crossotheca Höninghausi differ from the 

 microsporangia already described under the name of Telangiiirn 

 Scotti, which some have thought were the microsporangia of Lygiuo- 

 dendron, in several characteristics, especially in being free from one 

 another, and bilocular. 



A new species, Crossotheca Hughesiana, from the same locality 

 is next discussed, but here only the fertile condition is known. It 

 agrees with C. Höninghausi in all the chief structural characters, and 

 the two species have, as far as their microsporangia are concerned, 

 a very close affinity. The form and structure of this new fossil is 

 fully described and illustrated. 



The Author passes next to show that the morphology of the 

 fertile segments of Crossotheca is that of a pinnule modified for the 

 purpose of fructification. Figures are given illustrating a restoration 

 of the sorus. This fructification is contrasted with Asterotheca and 

 Dactylotheca. It is pointed out that two of the special features presented 

 by Crossotheca , are that the sporangia appear to have been quite 

 free from each other, or from any central point to which they were 

 mutually attached, and the fact that they are bilocular. In the latter 

 respect, it is difficult among recent plants to find a parallel case. 

 The origin of this bilocular microsporangium, regarded as due to 

 the coalescence of two sporangia, may be traced by means of certain 

 Palaeozoic fructifications, beginning with the fructification oiPalaeo- 

 pteris, and passing on to Telangium and Diplotheca. 



The Author nex enters on a füll discussion of the relationship 

 of the Pteridospermeae to other groups. The eight known species of 

 Crossotheca are regarded as probably all the male organs of this 

 group. The Devonian plant Archaeopteris, possessing a fusiform 

 sporangium, is also in all probability a Pteridosperm. The conclusion 

 is reached that so far we have no satisfactory evidence of the 

 existence of a Single true fern in Devonian times. Passing to the 

 Lower Carboniferous period, the fructifications Telangitun , Dactylo- 



