96 SCIENCE PROGRESS 



its chloroplasts has been shown definitely to be due to an excess 

 of xanthophyll (which is a prominent characteristic of the latter 

 group). We are thus left with the occurrence of oil as the 

 assimilatory product as the most prominent peculiarity of 

 Vaucheria, distinguishing it from other Siphoneae. In the genus 

 P hyllosiphon {^uscdiWom ii), which is parasitic in the leaves of 

 various plants, we have a form of doubtful position, although 

 certainly of siphoneous affinities. In this Alga, however, both 

 starch and oil occur side by side, and this may help to bridge 

 over the gap between the starch-producing Siphoneae and the 

 oil-producing VaitcJicria. 



In recent years a new siphoneous genus has been described 

 by Ernst (23) under the name of Dichotomosiphon ; this seems 

 to constitute a further connecting link between Vaucheria 

 and the Siphoneae (especially the Codiaceae among the latter, 

 cf. also Lotsy 46). Although Dichotomosiphon forms starch, it 

 has usually been placed side by side with Vaucheria (Blackman 

 and Tansley 6, Wille 70a) owing to the close resemblance 

 between the two in other respects. DichotoinosipJion (fig. 3, m) 

 consists of branched unseptate filaments, which are peculiar 

 in their dichotomous branching and in the possession of annular 

 membrane-thickenings (cf. fig. 3, m) at regular intervals, and in 

 both respects resemble very closely the filaments of many of the 

 Codiaceae. Zoospores have not been observed, but their place is 

 taken by large club-shaped spores (gemmae) formed at the ends 

 of special rhizoid-like branches (fig. 3, m) and on germination 

 giving rise to a new filament. We have only to imagine that 

 the terminal zoosporangium of a Vaucheria ceases to form the 

 synzoospore, and after a certain time germinates as a whole, to 

 obtain a structure very similar to the gemma of Dichotomosiphon. 

 This is almost precisely what occurs in certain species of 

 Vaucheria (e.g. V. piloboloides, cf. Ernst 25). Here the contracted 

 contents of the zoosporangium do not develop any cilia, but 

 become enveloped by a membrane before being protruded from 

 the sporangium (fig. 3, h), or they may even germinate from 

 within the latter ; in other cases the whole sporangium grows 

 out vegetatively to form a new filament (fig. 3, l) and this is 

 obviously almost identical with what we have in Dichotomo- 

 siphon} We have thus an almost complete transition between 



' It is also useful to compare these structures with the characteristic resting- 

 spores (akinetes) oi Piihophora (Cladophoraceae). 



