7 niet 
30 The Botanical Gazette. [January, 
and rhizobia than the normal root. It is also certain that 
there is greater cytoplasmic activity in tubercles than in the 
normal roots; this is shown by the greater abundance of al- 
buminous substances present and the greater prominence of 
the cytoplasmic granules* (Dermatosomen, Plasomen, etc.) 
(plate Iv, fig. 7). 
As already stated the dark tan and dark brown tubercles 
always contain the infecting nostoc. It is generally taught 
that algz can not develop in the dark. This is evidently not 
true as some of the nostoc-bearing tubercles are found as much 
as a foot below the surface of the soil. In fact tubercles 
wholly above the ground never contain the nostoc. That the 
nostoc is the cause of the development of the palisade layer 
is quite evident from their constant association. | 
The exact mode of infection is as yet undetermined. The — 
nostoc no doubt enters the parenchymatous tissue of the 
tubercle through a break in the dermal layer soon after it be- 
gins to form. Why the nostoc should take up a definite posi- 
tion in the parenchyma midway between the dermal layer and _ 
vascular system sheath is as yet unexplained. The paret-— 
chyma cells nearest the nostoc appropriate the extra nitro- 
genous compounds stored by the infecting symbiont, this 
produces hypernutrition of the incipient palisade cells which — 
elongate in a direction parallel to the easiest conductivity of 
nutritious substances, that is practically at right angles to the — 
vascular system. They serve a similar function to palisade 
tissues in other positions, as in leaves. Nostoc, so to speak, — 
takes the place and serves the function of chlosoplastids in 
true palisade cells. | a 
Nostoc is the only alga found in the tubercles. This is _ 
probably because it is more closely related to Schizomycetes- 
than Protococcus or Ulothrix. It is therefore better adapted 
.to lead a parasitic or a symbiotic existence. From a study 
of the infecting alga I conclude that it is Mostoc commun. ip 
Reinke placed it with Anabzna since he could detect 10 
gelatinous imbedding material. I found however that they 
were quite firmly united to each other and to the palisade cells 
by a gelatinous substance. Cells are spherical, loosely united, 
forming longer or shorter strings. Division takes place at 
right angle to nostoc chain. Sometimes a cell divides parallel 
ae 
* Contribution to the probabl biol f il. Torr. Bot. Club 
20: 339. Oct. 1893, P € biology of Plasomen, Bull. 2 el 
