383 
Physiology of the Saproleg niaceae. 
are so few or of such a grade of importance as to leave open the question 
of its rank. Later specimens of what seem the same fungus are received 
from other sources, and if all the new characters are also present in these, 
the evidence is considered sufficient to raise the fungus in question from its 
varietal position to the rank of a new species. 
This attitude fails to take into account one important consideration, 
viz. the question of the Constancy of Conditions under which these 
different individuals grew, or under which their constancy of character was 
tested. If, as has been brought out by recent investigators (Klebs, 5 06 ), 
the same conditions of nutrition, temperature, &c., produce always the same 
result, then the mere cultivation for a long period of time in and of itself is 
no test of species in the sense in which the term has been used. The problems 
of constancy include a trial of all the conditions. It would seem that we are 
driven to the same resources for accuracy in biological studies which are 
fundamental in physics and chemistry, namely, that of standards. Each species 
is recognizable, not only by certain prescribed or described characters, but 
in addition by these characters only when present under certain definite pre- 
scribed conditions. For example, Saprolegnia hypogyna is known by the 
liypogynous cell on the oogonial stalk and by certain other less important 
characters, all of which are present under the conditions of culture on a fresh 
fly, or more precisely in a definite per cent, leucin or haemoglobin solution, 
at a temperature, say, of 20° C., &c. Differently stated, this implies that 
the characters acquired in a new environment are persistent as long as the 
environment is persistent. It would seem necessary, then, in monographing 
such a family as the Saprolegniaceae to refer all the species to definite 
conditions, which should be uniform for all, and to determine in each case 
the variability to the extreme attainable limits ; we should then have 
attached to each description of a species an account of its variabilities. 
This view demands that comparisons with old descriptions should take into 
account the exact condition under which the plants were raised; a pro- 
cedure fraught, no doubt, with difficulties. Such a physiological basis for 
species seems to me the only starting-point in a revision of the relationship 
of the species of any family of plants. Whether it is practicable in all cases 
is another matter which cannot be discussed here. 
Another question which always arises in this connexion is the one as 
to whether species are distinct or whether there is a gradual blending of 
one species into another by means of intermediate shadings. In other 
words, whether we have new species formed by mutation or by continuous 
variation. It is not my intention to take sides. I merely wish to point 
out in what way my results touch upon the problem. As has been seen, 
some of the forms of Saprolegnia are very close to each other morpho- 
logically, and yet are physiologically distinct. This appears very strikingly 
in such forms of 5 . mixta as (H) and (F), along with the one that Klebs 
