552 
PROFESSOR- H. MARSHALL WARD OH THE 
is as well to bear in mind that until other forms have been investigated on the basis 
of the knowledge now to hand, there is always a certain amount of assumption in 
supposing that the tubercles of the roots of Leguminosas are all due to the Fungus I 
have described. Tschirch, like others, has by no means hesitated at assuming a 
common cause for these structures, however, and I am strongly of opinion that they 
are so far right, but that the causal agent is in all cases a Fungus. 
Tschirch regards the tubercles as normal structures, differing in form and anatomy 
according to two types. In Robinia and the majority of the Leguminosse the 
anatomy, &c., is so like what occurs in Vida Faba that we need not dwell on the 
details. He finds the development of the tubercles on the roots begins with that of 
the first leaves above ground, and traces coincidence of developments of tubercles and 
leaves onwards : at the close of the period of vegetation the tubercles become emptied, 
in whole or in part, as the seeds ripen. 
This emptying is considered to be due to the dissolution (and absorption into other 
parts of the plant) of the inner core of tissue containing the gemmules. Here comes 
in an important difference between our descriptions. Tschirch accepts a view, 
already propounded by Brunchorst,'"' that the gemmules belong to the root, and are 
formed by the protoplasm of the cells ; as will be shown shortly, the two writers 
differ somewhat in their interpretations of these bodies, but both employ the same 
name for them— Bacteroids. 
In Robinia Tschirch finds that the £! bacteroid tissue” is partially or nearly wholly 
emptied in autumn, and states that this is the rule. The phenomenon thus assumed 
to be a resorption of the bacteroids I take to be the escape of the gemmules from the 
tubercles; they are not absorbed, they are set free. 
Tschirch finds much starch in the young bacteroid tissue, but less as the cells 
become older; this is in accordance with my observations, but is subsequently 
explained differently. 
In annuals the tubers are found to attain a maximum of development, and then 
become emptied as the seeds ripen. 
Tschirch agrees with Brunchorst that the gemmules or “ bacteroids ” are not 
organisms, but bodies which arise by differentiation from the protoplasm of the cell. 
This view he bases on their “ development ” (though no account of their development 
appears), their fate (the assumed resorption into the plant), and the normal occurrence 
of the tubers in all genera of the strictly defined group Leguminosse. Tschirch also 
adds that every attempt to cultivate the “ bacteroids ” has failed, and that they could 
not enter the tubercles, because the latter are covered by cork. 
Reference is made in a footnote (p. 67) to the observations of Woronin and Frank 
as to doubtful development of the gemmules (bacteroids) in artificial cultures. Why 
it should be assumed (p. 67) that “ im vorliegenden Falle hat man es, wenn 
* ‘ Berichte der Deutschen Botanischen Gesellscliaft, 1885,’ p. 241; and ‘ Untersucliungen aus dem 
Botanischen Institut zu Tubingen,’ vol. 2, Heft 1, 1886, p. 151. 
