i5 
Davis . — Spore Formation in Derbesia. 
characters of a primitive ancestry from forms allied to the flagellates. 
In other words the periods of sporogenesis and gametogenesis are regarded 
as periods in the life-history when the algae return for a short time to con- 
ditions similar to those of their ancestry, and the structure of these motile 
cells may be expected to give important evidence of the morphology of the 
flagellate-like types from which various classes and orders of the algae may 
have been derived. 
This view unquestionably has much of value both in practice and 
theory. However, its application may be carried to extremes, and classifica- 
tions based chiefly or wholly on the zoospores and motile gametes are 
likely to prove artificial and open to criticism, after the manner of many 
systems which have been proposed on single characters. We have in the 
Siphonales an assemblage of algae which illustrate very clearly the danger 
of attaching too much importance to these reproductive cells as taxonomic 
characters, and Derbesia is of especial interest as a test case. 
Blackman and Tansley (’02) in their revision of the classification of the 
green algae have removed Botrydium and Vaucheria from the Siphonales 
because the motile reproductive cells do not conform in structure to the 
zoospores and gametes typical of this group. Although they still retain 
Derbesia because of its general resemblance to Bryopsis their opinions are 
strongly expressed in the following remarks (p. 25 of the reprint) : 
‘ Though the vegetative characters of this genus suggest a position among 
the lower Siphoneae in the neighbourhood of Bryopsis , yet the peculiar 
zoospores, recalling those of Oedogonium , probably indicate that its affinities 
are widely different. Our present information is not however sufficient to 
justify the removal of the genus from its traditional place.’ 
An analysis of the process of spore formation in Derbesia shows at once 
how unsafe it would be to conclude that the peculiar structure of its 
zoospores has deep phylogenetic significance. The early stages of spore 
formation are those common to all of the Siphonales (including Vaucheria ) ; 
the sporangium becomes filled with multinucleate protoplasm. Then in 
Derbesia there begins the peculiar process of nuclear differentiation and 
the degeneration of the greater number of smaller nuclei. The process of 
nuclear degeneration is very exceptional, although finding its analogy during 
oogenesis in Vaucheria , Saprolegnia , and in the Peronosporales. The 
result is the development of a set of extraordinarily large nuclei which 
become the centres of spore formation. The kinoplasmic activities con- 
cerned with these large nuclei produce blepharoplasts of remarkable form 
and size, capable of developing and bearing a great many cilia instead 
of the typical number two. 
There can be little doubt that the peculiarities of the zoospore of 
Derbesia are later developments in phylogeny concerned with the excep- 
tional features of the process of sporogenesis, and that its structure is not 
