270 THK .TOURNATi OF P.OTANY 



seems to be no valid reason why the British and Irish plants 

 should not be reduced to one species, A. filiculoidcs. 



The problem arose in 1912 when Ostenfeld diagnosed the 

 Norfolk plant, and his opinion had to be harmonized with the 

 generally accepted belief, based on high authority, that the plant 

 at Woodbastwick was A. caroliniana. The easy and natural 

 solution of the difficulty was to assume that two species were 

 present, which seemed not improbable, because the material 

 grouped itself into two well-defined types : one distinguished by 

 its pale green leaves with hyaline border, standing at a wide angle 

 with the stem, the fronds riding high and very buoyantly on the 

 water ; the other having its dull olive-green leaves edged with 

 red, closely imbricated, the fronds being smaller and more closely 

 appressed to the water. An attempt to make the smaller plant 

 A. caroliniana failed ; it was always barren, and, after a prolonged 

 study, the opinion was formed that it was an immature state 

 which eventually passed to the adult fruiting state of A.filiculoidas, 

 the altered angle of the leaf-lobes and the increased buoyancy of 

 the fronds being due to inflation of air cavities in the basal 

 tissues. Direct evidence is not available to show {a) that these 

 small immature plants develop the fruit of A. filiculoides in due 

 course ; (b) at what stage in development the marked inflation of 

 the air cavities takes place. In culture tanks the plant deteriorates, 

 and observation of isolated fronds in statu natures, has failed, 

 owing to fluctuation of water-level at Woodbastwick from tidal 

 causes. No attempt has been made to grow it on uninfected 

 waters, for sentimental reasons. While it is desirable that these 

 two details should be cleared up, it may be stated in support of 

 the relationship of the two forms that (a) they usually occur 

 together in Norfolk, and have been received mixed from other 

 stations. A vast quantity of the immature state was noted in 

 Ean worth and Walsham Broads in October, 1913. The distri- 

 bution of Azolla in the Bure Valley by flood took place in August, 

 1912, just at the period when fruit was approaching maturity. 

 Eeproduction ])y spores was to be expected, and would account 

 for the large proportion of immature material present at the later 

 date. (6) The smaller plant has repeatedly been diagnosed as 

 A. caroliniana, but whenever fruit has subsequently been produced, 

 it has proved to be that of A. filiculoides, and there seems to be 

 no clear reason for assuming that two species are involved. 



In considering possible causes for the existing confusion, 

 Campbell's history of the plant in America seemed to offer a 

 useful clue: — "^4. filiculoides is confined to the western part of 

 America, being reported from Chile to California at least and 

 probably beyond. Until very recently American botanists con- 

 founded this species with A. caroliniana of Eastern America, and 

 in the Botany of California only that species is mentioned. I 

 have examined material from various parts of California, and in 

 all cases the plants were undoubted specimens of A. filiculoides " 

 (Ann. Bot. vii. 155). 



The introduction of material incorrectly named, and its distri- 



