24 T. S. COBBOLD ON THE EMBRYOLOGY OF ACHIMENES PICTA. 



Btatus of the frond-bearing, spore-2)rodiicing asexual generation in 

 Cryptogams will have to undergo revision, and perhaps reversal. Be 

 that as it may, the position assumed by Hofmeister, Pfeffer, and 

 others, seems to be this : — The endosperm, although not an inde- 

 pendent structure, and formed within the embryo-sac (where, as we 

 have seen, it is often multicellular or parenchymatous before the 

 essential act of imj)regnation takes place) is strictly comparable to 

 the free, growing jDrothallium of a fern or other Cryptogam. The 

 embryo-sac corresponds to the macrospore, and the pollen grain to 

 the microspore of " heterosporous vascular cryptogams." As the 

 nucleus or tercine j^i'oduces the embryo-sac, this tercine, is 

 deemed the equivalent of the macrosporangium, and so forth. The 

 analogies extend much further, but I have probably adduced 

 sufficient to show the nature of the data on which Sachs and 

 others have founded the proposition that " a concealed alterna- 

 tion of generations exists in the seed of phanerogams." The 

 phenomena of embryonal development, as witnessed in Cycads and 

 Conifer^e are, as they tell us, strikingly different from what obtains 

 in Monocotyledons and Dicotyledons, and since it is alleged that the 

 prothalloid endosperm in these orders actually produces achegonia- 

 like organs, these groups of plants seem to occupy a transitional 

 position between Cryptogams and Angiosperms generally. It 

 happens, likewise, that according to Pfeffer' s investigations (quoted 

 by Sachs) there is a still more striking connecting link arising out 

 of the unique fact that amongst Cryptogams the development 

 of a suspensor itself may be actually witnessed in Selaginella 

 (Lycopodiacece). 



