140 
E. IV. Thomas. 
the country had hardly begun and that care must be taken to secure 
its adequate prosecution. 
The President then proceeded to consider certain of the 
Australian plants from the phylogenetic aspect, in particular the 
Pteridophytes, and finished with a very illuminating presentment 
of the difficulties of evolutionary investigations. 
The peculiar Australasian genus Phylloglossum loses much of 
its phylogenetic interest as an archaic type, if we subscribe to the 
view which sees in the “ protocorm ” a biological adaptation to 
annual growth. Professor Bower shows, however, that we can 
hardly avoid this conclusion if we compare Phylloglossum closely in 
its habit and mode of growth with what is now known concerning 
the various species of Lycopodium in particular with L. inundatum 
and the Australian species L. laterale, the former of which produces 
small protocorms each year, while the latter has an embryonic tuber 
of exceptional size. Phylloglossum would seem, therefore, to have 
carried this tuberous habit to its fullest development and both starts 
and continues life on geophilous lines. 
Professor Bower dwelt at some length on the peculiarities of the 
Australian members of the ancient type of Ferns so well represented 
in the Australian flora, and showed that the most perfect demon¬ 
stration of their archaic character had been built up on a New 
Zealand fossil Osmundites which furnished the first member of the 
brilliant series of papers on Fossil Osmundaceae. The monotypic 
New Zealand Fern Loxsoma shows characters shared by the isolated 
Juan Fernandez form Thyrsopteris and by the recently discovered 
Jurassic fossil Stachypteris Halli. This curious distribution of 
characters would seem to indicate that they represent relics of a 
once widespread Filicinean condition. Professor Bower then drew 
attention to the fact that the isolation of the Hydropterid condition 
bids fair to be bridged by Professor Lignier’s discovery of the 
heterosporous form Mittagia. 
The dfficulties of phyletic interpretation which present them¬ 
selves even among Ferns—in many ways one of the eaziest groups— 
was shown to be largely due to the prevalence of parallel develop¬ 
ment in phyla of distino^origin. This difficulty is very acute among 
the Angiosperms, in which no fossil appeal is as yet possible, and in 
which the “ irritating sameness ” of their essential features on the 
one hand and the great adaptability of the multitude of forms on 
the other, render phylogenetic unravelling almost impossible. The 
great need here is for a widening of the criteria of comparison which 
