io6 SCIENCE PROGRESS 



are also represented. The series is much more definitely circum- 

 scribed than is that of the Isokontae, although this may possibly 

 be due to lack of knowledge on our part. It is interesting to 

 note that the transition from Flagellate to algal characteristics 

 took place at rather a different point in the Heterokontae than 

 it did in the Isokontae. For the algal groups of the former do 

 not include any motile unicellular forms comparable to Chlamy- 

 domonas and its allies, and even the TctrasporaAWiQ Chlorosaccus 

 is still a Flagellate in all essential characters. 



Since the establishment of the group of the Heterokontae it 

 has become customary (Blackman and Tansley 6, Oltmanns 53) 

 to include in it the genus Botrydium^ a small ccenocytic form 

 inhabiting the damp mud round the edges of pieces of water. 

 It shows some points of resemblance to the genus Protosiphon 

 (cf above), with which it was formerly confused (Klebs 40), 

 Like this genus it consists of an overground vesicular portion 

 and a colourless subterranean rhizoid, which is, however, richly 

 branched (fig. 4, f). It agrees with the Heterokontae in forming 

 oil as the product of assimilation and in possessing zoospores 

 showing considerable resemblances to the Heterokontan type. 

 These zoospores are produced in immense numbers by sub- 

 division of the contents of the vesicular portion, and are liberated 

 through a small aperture formed at the apex of the latter. 

 They are provided with a single cilium (perhaps accompanied 

 by another very short one?; cf footnote on p. 104) and have 

 two chloroplasts, as is frequently the case in the zoospores 

 of Conferva and its allies. There is also resemblance to the 

 latter in the fact that Opiiiocytiimi (Confervales, fig. 4, o) shows 

 a marked ccenocytic tendency. These indications of affinity are, 

 however, all somewhat remote, and a difficulty is also to be 

 found in the fact that the chloroplasts of Botrydiiim are stated 

 to have pyrenoids (not forming starch) in young stages (Klebs 

 40, p. 224), although they disappear later on This latter feature 

 may possibly point to derivation from an Isokontan ancestor, 

 and the ccenocytic character of Botrydiiim of course suggests 

 an affinity with the Siphoneae. Wille (70a, p. 51) classes 

 Botrydiiim together with Protosiphon in a separate family, the 

 Hydrogastraceae, which he regards as a connecting link between 

 his Protococcaceae and Valoniaceae. Apart from their general 

 organisation the two genera do not, however, appear to have 

 ' Regarding Vauchcria and Dichotoiiiosiphon^ see pp. 93 and 96. 



