'fe. 



'PPosiJ* 



I 



^ 



as 





^ ^"glens fs,j ^ 



ucheri 



a and I 



themsel 



ves iit, 



i 



e 



p-acewitht: 

 orpuscles of 

 of a 



paleg 



ter limits itsd 



:nules appear ii 



(Fig. 82, n, I 



s to form 







aive off latos. 



^■ 



alsi. Atotk! 

 late most pk«' 



ifferent 



1 



: por 



rtioD 



1 



i 



nlarges acf ^ 

 3 some ki"J 



P 



ecimens «^ 



f a very 



brig" 



or -<;; 



inu 



ent 



a 



:02') 



rjy^ BEGINNINGS OF LIFE. 



419 



green colour (<?, ^.) These filaments sometimes appear 

 to spring at once from the outer layer of the filament, 

 and sometimes from an expanded base, also having 

 a greenish colour, which soon becomes surrounded by 

 a dark brown decolourized- zone (^, e .) The zone 

 oft^n persists long after the Desmid itself has dis- 

 appeared. At other times I have seen attached to the 

 same kind of base (on specimens of Cladophora) a large 

 ovoidal and perfectly hyaline envelope containing in its 

 centre an aggregation of bright green algoid spherules 

 {a\ each of which seems to be capable of enlarging and 

 transformino; itself into an Astasia. Diatoms also ap- 



pear to grow from the surface of the filamentous Alg-^, 

 to' which they are afterwards seen to be attached by a 

 hyahne tubular pedicle; and it is this mode of origin^ 

 apparently, to which Dr. Gros referred when he said ' : 

 ^C'est une chose soupgonnee sinon reconnue par tous 

 Ics observateurs que des vesicules vertes se font jour a 

 travers le tube des Conferves, et se convertissent en 

 Navicules/ Although many Alg^ present phenomena 

 of this kind, in none are they more striking than in 

 certain specimens of Vaucheria. Filaments of this 

 weed which but a few days before — when placed for 

 observation in a watch-glass — may have showed neither 

 Diatoms nor Desmids upon their surface will at times 

 become crowded, both inside and out, with a rich va- 

 riety of both of these modes of growth. And although 

 it is quite possible that in all such instances the 



- • . /^ Loc. cit., p. 316. 



E e 2 



