LIXNEAN SOCIETY OF LONDON. 4 1 



found a good specimen of MeduUosa antjUca, which enabled him to 

 add to our knowledge of the roots, and to describe the structure 

 of the sieve-tubes, a point rarely observable in petrified plants, 

 (13, 1903). In 1901 Arber named and described a sihcified tree- 

 trunk from Tasmania, of Tertiary age, recorded by Sir Joseph 

 Hooker more than 00 years before (17, 1901). A paper on new 

 species of Lar/enostoma is of special interest from the fact that in 

 one of them the cupulate seeds were still attached to the rhachis 

 of the frond (20, 1905). 



Arber's British Museum Catalogue of the Glossopteris Tlora, an 

 important systematic work, contains an account of the probable 

 sporangia associated with the scale-fronds of Glossopteris, whicli he 

 had described more fully in a separate paper (22 & 23, 1905). 

 This was the first time that any light had been thrown on the 

 reproduction of this important fossil genus. 



In a valuable theoretical essay on the past history of the Ferns, 

 Arber proposed the class-name Primofilices for the Palaeozoic 

 stock from which the Mesozoic Leptosporangiatae were derived, 

 the BotryopteridesB (in the wide sense) being the best known 

 family within the group. Tlie name Primofilices has often been 

 misunderstood; it was not intended to imply the primitiven.ess of 

 the group, but rather its antiquity, as being characteristic of the 

 Primary Rocks. Rightly interpreted the designation serves a 

 useful purpose. In tliis memoir Arber expressed a judicious 

 scepticism as to the supposed dominance of the Eusporangiate 

 Perns in the Palaeozoic ; he further pointed out that the resem- 

 blanre between the sporangia of such Perns and the pollen-sacs 

 of the lower Spermophyta, may well be a case of homoeomorphy 

 rather than a mark of affinity, a suggestion whioli now appears 

 worthy of serious consideration (21, 1906). 



In the well-known memoir by Arber and John Parkin on the 

 Origin of Angiosperms (29, 1907) the idea of a relation between 

 Angiosperms and Bennettiteae, already suggested by Wieland, is 

 worked out in detail and elaborated into a definite phylogenetic 

 theory. " It appears to us," they say, " that although the direct 

 ancestors of the Angiosperms are as yet unknown in the fossil 

 state, this line of descent can now be traced back to the great 

 group of Mesozoic Cycadophyta, and to a hypothetical race of 

 plants nearly related to the Bennettiteae" (p. 30). Wieland 

 himself expressed a very favourable opinion of the pajjer. Shortly 

 afterwards he discovered in his WUliamsonia mexicana a type of 

 stamen which almost exactly realises that postulated by Arber 

 and Parkin in their reconstruction of the hypothetical Hemi- 

 angiosperm. 



The present writer, who has maintained essentially the same 

 hypothesis of the origin of Angiosperms, was first put on the 

 track in a conversation with Arber. Mrs. Az'ber tells us that 

 Professor von AV^ettstein of Vienna " pro[)Osed and arranged the 

 translation of the Origin of Angiosperms into German." " He 

 wrote concerning this paper: ' As I have been occupied for some 



