130 
FRANK DUNN KERN 
one but is not the most important one. The bacteria are notable 
organisms which raise such questions. But the genetic relationship 
is of greater significance. It is a common view that the forms mor- 
phologically alike must have had a common origin and that physio- 
logically specialized forms have resulted by progressive variability on 
a special host. Nutritional adjustment and other interacting factors 
have been suggested as associated with the variations but no demon- 
stration up to the present has made the matter at all clear. It has been 
assumed that continual specialization will lead eventually to structural 
differentiations. There seems to be nothing in the methods of descent 
which would prevent the carrying forward of any variation once es- 
tablished. It is rather a question of how the physiological relations 
between host and parasite or the environment might operate to bring 
about a diversity of form and structure. 
Very little is known of the possible effect of the ordinary factors of 
environment upon the development of a parasite. There is some 
evidence regarding effects upon distribution. After a detailed com- 
parison of the geographical distribution of hosts and parasites in one 
group of fungi, the writer came to the conclusion that the environment 
had little if any influence at all, but such apparently is not true for all 
kinds of parasites. Ward has shown that external influences may 
undoubtedly exert important effects. Too high a temperature during 
incubation, starving the host of carbon dioxide or salts, or heating or 
cooling the roots may ruin a fungus and stop infection. Salmon has 
also carried out investigations to show that external as well as internal 
features are concerned in determining whether parasites may attack 
and continue development. 
As a conclusion we may consider, with brevity, parasites as an aid 
in determining genetic relationships in their hosts. Cobb has written 
on this subject in a very enthusiastic way, pointing out that parasites 
may be used (i) as an aid in discovering specific and generic relation- 
ships, (2) in following metamorphoses, (3) in physiology, (4) in 
chemistry. I am particularly interested in the first of these. I trust 
I may be pardoned for drawing my examples from a group in which I 
have long had a special interest. A recently published result brings 
this phase of the subject to mind. The members of the genus Gym- 
no sporangium were long supposed to alternate solely between the 
juniper and apple families. Upon the morphology of a spore-form 
found upon a member of the rose family I suspected its connection to a 
