140 Transactions of the Society. 



an extensive agreement, but the stem retains a " Cryptogamic " 

 character in the presence of strands of centripetal wood. Kenault 

 himself regarded the family as standing, on this account, a step 

 nearer to the Lycopods ; in the light of our present knowledge we 

 should rather compare them with the Pteridosperms. However that 

 may be, the extraordinarily perfect knowledge of their anatomy, 

 which we owe to Renault and his colleague, has given the group a 

 crucial significance in the discussion of G-ynmospermous affinities. 



5. Fossil Pollen-Grains. 



In the course of his work with Brongniart on the wonderfully 

 preserved silicified seeds of Autun and St. Etienne, Renault dis- 

 covered the constant presence of a pollen-chamber— i.e. a definite 

 excavation in the tip of the nucleus, adapted for the reception of 

 the pollen-grains. Guided by his observations on Palaeozoic seeds, 

 he was able to detect the same organs in those of their nearest 

 living allies, the Cycads, in ignorance of the fact that this discovery 

 had already been made by Griffith, thirty, or perhaps even forty, 

 years before. In the pollen-chamber of the various fossil seeds 

 the pollen-grains, which were to effect their fertilisation, are often 

 found in an excellent state of preservation, as we have already 

 seen in the case of Cordaianthus. In this position especially, 

 though also when still contained in the microsporangia, the pollen- 

 grains proved, in many instances, to have a multicellular structure. 

 This at least was the interpretation which Renault put upon the 

 septate appearance which they present. Attempts have been 

 made to show that the apparently multicellular structure is due 

 merely to the occurrence of folds in a contracted inner membrane, 

 but recent work has completely confirmed Renault's view, and in 

 many cases, at any rate, there appears to be no doubt that a body 

 consisting of several cells was really present within the pollen- 

 grain. The analogy with the male prothallus and antheridium of 

 a cryptogamic microspore is manifest, and was at once perceived 

 by Renault. In the Gymnosperms of the present day, the cor- 

 responding structures, though present, are less conspicuously 

 developed. 



This discovery, so interesting in itself, led its author to form 

 certain hypotheses as to the method of fertilisation, which are of 

 remarkable interest in the light of subsequent results. In ex- 

 amining a seed named JEtheotesta, probably belonging to a Pterido- 

 sperm, Renault, in 1887, observed the minute structure of the 

 multicellular pollen-grains in the pollen-chamber. He found that 

 both in the outer wall of the grain, and in the septa between its 

 cells, there were perforations. He suggests that the latter may 

 represent the traces of insertion of multiple pollen-tubes, or — and 



