178 THE JOURNAL OF BOTANY 



direction of osmosis in the case under consideration lias yet to be 

 confirmed, but the structure of the hyphai as a network of long 

 capillary tubes may easily allow, by a very similar physical process, 

 a transference of food-substances in the hyphal filaments, on account 

 of the uninterrupted capillary attraction of the hyphfe. 



If it is impossible to speak with certainty of the importance or 

 otherwise of the contact of the gonidia and the hj'^phfe, it is at least 

 certain that the penetration of the gonidial cells at a later stage is 

 attended with obviously fatal consequences to them. Their contents 

 become deformed, contract, and finally the algal cell entirely perishes. 

 As a consequence of this intracellular action, the algal cells some- 

 times cut off daughter cells, which appear free from haustoria. 

 Perhaps by this method some of the algal cells, which are found in 

 comparatively good condition, having freed themselves from the 

 haustoria, continue their growth ; or, possibl}^ by some other method 

 the same result is attained. In spite of the assertions of Hedlund, 

 Schneider, and Peirce, the algal cells do not always free themselves 

 from the invading haustoria by fission into daughter cells, for the 

 hyphee, rapidly spreading out inside the mother cells, frequentl}'^ 

 destroy many of the gonidial daughter cells. The haustoria having 

 penetrated into the gonidial cells appear devoid of cell-wall, perhaps 

 owing to the action of the living protoplasm of the invaded gonidia. 

 The protoplasmic hyphal filaments, feeding on the gonidial protoplasm 

 form a network at its periphery, and gradually forcing their way 

 form local swellings, or beaded distentions which gradually collect 

 together and perhaps mark a definite stage of the development of 

 the fungoid element in its intragonidial life. As the gonidial proto- 

 plasm disappears, the hyphal formations continue their development 

 further and further, and when the contents of the gonidia have been 

 entirely absorbed, the plasmatic branches from the hypha? are ex- 

 tended until they fill the wliole gonidial envelope. This phase of the 

 development of intragonidial hypha? seems to me to correspond with 

 the "pale gonidia" described above. It cannot easily be stated 

 exactly what further happens to these h^q^hal formations enclosed in 

 the cell-wall of the destroyed gonidia. 



I examined these gonidial envelopes, the contents of which con- 

 sisted of numerous separate and disconnected masses similar to those 

 contained in the interiors of the "pale gonidia." Some of these — 

 generally speaking, those which were not deformed — were found lying 

 in the perforated parts of the gonidial envelope, possibly as the result 

 of pressure from the cover-glass. It is possible that such formations 

 present us with some hitherto unexplained stage of fungoid develop- 

 ment, or perhaps they, having no other function to fulfil save that 

 of haustoria, become themselves covered with a thick envelojje on 

 reaching the inside of the gonidial oell-wall. Just as, in my opinion, 

 occurs to the haustoria in other cases, when the gonidial protoplasm 

 has been disorganised or destro_ved. As the outcome of the above- 

 described stage in the development of the haustoria we are in a 

 position to observe such appearances as that of a gonidial envelope 

 absolutely packed with closely interwoven hyphie. The superficial 

 fusion of the intragonidial hyphal growths with the external hypha^ 



