1893.] Root Tubercles of Leguminose. 787 
lenticels consist of the usual loosely connected, rounded cork 
cells and are always located above the vascular bundles. They 
no doubt serve as a means for the interchanging of gases 
between the exterior and interior by way of the vascular 
bundles. Frank maintains that the cork layer as well as the 
infected area have intercellular air spaces for the interchange 
of gases. As already mentioned the infected area consists of 
closely united firm walled cells, such as are usually found in 
meristematic tissues. Furthermore the infected area is entirely 
cut off from the cork layer by means of the cambium which 
certainly has no intercellular passages. Hence it would be 
rather difficult to see what function intercellular spaces would 
play in the infected area. Tubercles possessing lenticels no 
doubt give off considerable gas. I have noticed that the 
tubercles of a growing bean plant when placed in water would 
have a glistening appearance due to a thin layer of air or some 
gas separating them from the water. After a time very minute 
gas bubbles would form on the surface of the tubercle. As to 
the source and nature of this gas I am unable to give any 
satisfactory explanation. It is very likely dependent upon the 
largely increased metabolic processes going on in the tubercle, 
-Frank has made some experiments on the subject without 
however, coming to any definite conclusions. 
Starch is usually present to some extent in tubercles, especially 
in those infected by Rhizobium mutabile. There is some in 
~ tubercles of Amphicarpaea comosa, little or none in tubercles 
of Phaseolus vulgaris and Robinia pseudacacia. The starch is 
always found in the noninfected meristem tissue especially the 
cork phellogen and next to the infected area. This is simply 
stored starch like that found in other parts of the root. 
Frank maintains that there are two varieties of Rhizobia 
always to be found in two different kinds of tubercles of Pisum 
sativum. One variety is said to have the power of producing 
within itself highly refractive amyloid bodies closely related 
to amylo-dextrin found in some of the Florideae. Moeller 
agrees with Frank in regard to the presence in some Rhizobia 
of the highly refractive bodies but maintains that they consist 
of some fatty derivative, as cholesterin. Both agree that these 
