100 A SYSTEMATIC AND STRUCTUBAL ACCOUNT OF AVEAINVILLEA. 



of course necessary, since the frond-filaments at all events are very 

 straight, and run nearly parallel to each other. We have been 

 much struck by the appearance in this connection of the fossil stem 

 of Nematophycus. In describing this type, Mr. Carruthers, besides 

 showing that it is truly algal in nature (and not a coniferous trimk, 

 as Sir William Dawson had, absurdly enough, declared it to be), 

 pointed to PenicUlus and Halimeda, to this group that is, as 

 exhibiting now a similar structure to that of which we have the 

 remains preserved in Neinatophycus. Graf Solms (* Palaeophytologie,' 

 pp. 46 & 86) seems to be inclined to regard Neniatoj^hycus as more 

 nearly related to FncacecB than Siphonea. Against this view there 

 is a good deal to be said, notably the open continuous structure 

 of the tubes, — exactly like those of Udotea, for example, — and 

 certainly unlike anything known to us in FucacecB. The late 

 Prof. Dickie has written as a note on a specimen of Avrain- 

 villea nigricans (his FMpilia Piawsoni) in his herbarium, "In 

 structure nenT Nematophycus j' '' The following reasons, however, 

 appear to point to a closer resemblance in structure between 

 Udotea and Nematophycus. The small filaments that wind about 

 among the large axial filaments of Nematophycus were compared by 

 Mr. Carruthers with the filaments of Halimeda, for example, which 

 branch ofl' and pass outwards, forming the cortical layer of that 

 alga. They resemble much more closely the tenacular filaments of 

 Udotea, and there is in some cases a very strong resemblance 

 between the frond-filaments of Udotea and the axial filaments of 

 Nematophycus. It appears very reasonable to assume that the 

 small winding filaments of Nematophycus are tenacular in function, 

 whether they terminate in actual tenacula or not, since the axial 

 filaments are comparatively little interwoven, and since we have no 

 reason, moreover, to assume that there was an incrustation in 

 Nematojjhycus. What else, we may ask, held together the great trunk 

 of Nematophycus, composed of filaments little interwoven ? There 

 is, at all events, no other explanation at hand of this mechanical 

 problem. It appears therefore extremely probable, after a minute 

 study of the group of Udotece, that we have in this group the living 

 representatives of a gigantic, tree-like, siphoneous Alga, which 

 inhabited the seas of the Devonian age. 



As for the reproduction of Avrainvillea, we have few observations 

 to submit, and these are not very conclusive. Fig. 5, Plate 288 

 represents what we take to be the terminal joint of one of the 

 moniliform frond-filaments of A. hngicaulis. It has become sepa- 

 rated from the rest of the filament, and the slender moniliform 

 tube attached appears to be an outgrowth from it. It was seen in 

 the living state by one of us in Grenada. Attempts were made 

 to procure the further growth of the slender filament, but these 

 were not successful. Swollen terminal and interstitial joints are 

 not uncommon in A. hmyicaulis, but these are not, so far as we 

 know, ever cut ofi" by transverse septation. They recal the torulose 

 cells of a SaproUynia, and possibly serve some such reproductive 

 function. At the same time, in the absence of further evidence it 

 is equally open to us to assume that the cell in question became 



