298 Ulotrichales 



scarcely to be separated from it. Chxtonema is a reduced type with a more or less 

 endophytic habit. Pseudochsete is either a reduced form or one which has permanently 

 remained in the juvenile state, and the two known species of this genus greatly resemble 

 the young states of species of Stigeoclonium. One species of Acroch&te (A . parasitica) is 

 a partial parasite in Fiicus. 



Stigeoclonium is largely a vernal type with a decided preference for the well-aerated 

 water of streams and springs. In this connection the growth of Stigeoclonium tenue on 

 living fish in stagnant waters, as mentioned by Hardy ('07) on Carassius auratus in a 

 Melbourne fish-pond and by Minakata ('08) on the small fry of Haplochilus latipes in 

 a tiny bog-pool at Tanabe in Japan, is of much interest, since by securing a foot-hold on 

 a moving substratum the Alga has been able to obtain, even in stagnant water, conditions 

 which approximate to those under which it more normally lives. Hardy ('10) has also 

 shown that the presence of the Alga on the living fish is in some cases dependent upon 

 the preceding growth of a species of Saprolegnia, in the mycelium of which the zoogonidia 

 of the Alga become enmeshed. 



A ' palmel la-state ' occurs in certain species of Stigeoclonium and may be induced by 

 cultures under subaerial conditions, in solutions of high osmotic pressure, and in sea- 

 water (Livingston, '05). 



The experimental work of Klebs ('96) on Stigeoclonium, by which he endeavoured to 

 show that the stimuli which caused the formation of asexual cells (zoogonidia) or sexual 

 cells (gametes) were distinct, and depended upon definite changes of environment, requires 

 repetition, since his conclusions have not been confirmed, and other observations, such as 

 those of Welsford ('12) on Trichodiscus, do not support them. 



Sub-family ULVELLE^E. In this sub-family the thallus is more or less 

 discoidal, flat, lenticular or cushion-like, and usually epiphytic. In some 

 forms (Pringsheimia, Protoderma) it is only one layer of cells in thickness, 

 but as a rule it is several layers thick in the middle and becomes gradually 

 thinner towards the periphery, which is always one-layered. In the two 

 genera Ochlochsete and Ch&tobolus (fig. 193) many of the cells are provided 

 with long tubular setas or hairs. In Arthroch&te there are also long hairs, 

 which are several times septate near the base, the basal wall always occurring 

 at the junction of the seta and the supporting cell. The other genera of the 

 Ulvellese are without seta3. 



The discs really consist of recumbent branches which have become more 

 or less completely concrescent, and in Ulvella they may attain a diameter of 

 1*5 mm. In Chtetobolus (fig. 193 D and E) all evidences of a branch-system 

 have disappeared and the cells divide in all directions of space, giving rise to 

 a hemispherical or sometimes almost a globular thallus. The thallus of 

 Pseudopringsheimia is also very thick and cushion-like ; it increases in 

 diameter by the vertical and tangential division of its peripheral cells, but 

 the increase in thickness is due to numerous transverse divisions parallel to 

 the plane of attachment, a section of the median part of the disc having the 

 appearance of a concrescent mass of vertical filaments. 



There is one parietal chloroplast in each cell, disc-shaped or somewhat 



