in the Sapro legnieae. 451 
of such great importance, merely because my observations 
contradict the conclusions which he drew from his own. 
His account is as follows : ‘ In Saprolegnia mixta , the 
young oospores have frequently a second nucleus near the 
point of contact of a fertilizing tube. No opening can be 
seen in the tube, and it would in any case probably be very 
difficult to prove its presence even were it there. But 
a permanent opening does not appear to be necessary for 
fertilization to take place in such cases. No one, for 
example, has been able to determine a permanent opening 
at the end of the pollen-tube of the Angiosperms, where 
fertilization undoubtedly takes place’ ( 95 , p. 635). True, but 
in Angiosperms (1) the pollen-tube becomes permanently 
attached to the embryo-sac, as is the case in other sipho- 
gamous fertilization ; (2) the pollen-tube loses its turgescence. 
Neither of these is the case in the Saprolegnieae ; as De Bary, 
Marshall Ward, Humphrey, and I have seen, the fertilizing 
tube may grow up to the naked oospore, and even dent it, 
but it then glides off from its surface and grows past it with- 
out losing turgescence. Trow’s conclusions appear then to be 
based, so far as Saprolegnia is concerned, on forms in which 
the final fusion of the last two nuclei left in the oospore- 
origin is habitually delayed to the same extent that I found, 
exceptionally, in a form free from male organs. 
In Achlya , he writes in ’99 (p. 159), ‘large numbers of 
sections were examined to try and elucidate the mode 
of fertilization, but only one section, that reproduced in 
Fig. 45, appeared to be capable of throwing any light on 
the actual process. In this case it was possible to trace the 
fertilization-tube without a break into an egg which was 
already surrounded with a thin membrane.’ The figure as 
it stands appears, at first sight, to be conclusive. But there 
are points about it that make me hesitate and finally reject 
its evidence. The cell-membrane is represented on the 
oosphere and the antheridium at the surface of the oangial 
wall, but the tube itself is represented as a grey homogeneous 
mass, with neither plasmic network, granules, nor nuclei, and 
