I conclude from the preceding experiments that the cenocytes, which have been regarded by previous investigators as glands, secrete oxidizing enzymes. I do not know whether this is their only function, but it is certainly one of them. At any rate they do not secrete a fat-splitting enzyme. Since these cells, hanging loosely to the tracheæ, lie free in the blood, the enzymes which they secrete may activate the oxygen of the body towards combustion. That the cells actually secrete is indicated by the fact that numerous observers, myself included, have detected microscopical exudations around the periphery of the cytoplasm, especially at times when the nucleus is greatly ramified, and therefore manifesting its great activity. Exactly what relation the \l=ce\nocytes bear to the tracheæ, I am unable to say. I saw no definite attachment, but am inclined to believe that a relation exists and that through this the oxydases, one branch of the group of oxidizing enzymes, are able to get their molecular oxygen with the formation of peroxides. It must, moreover, be remembered that Wheeler, in 1892, found the \l=ce\nocytes of phryganeid larvæ to be provided with delicate processes which are attached to the tracheal hypodermis. That may, of course, be simply a means for attachment and have nothing to do with the passage of oxygen from one to the other. The location of the \l=ce\nocytes may be purely due to the absence of certain mechanical forces. Wheeler found that "the \l=ce\nocytes originate by delamination or immigration from the ectoderm, just caudad to the tracheal involutions and after their differentiation from the primitive ectoderm never divide, but gradually increase in size." Since they never divide after differentiation, the mechanical forces of cell division being absent, they are not able to be carried far from their starting point. Moreover, since there are no other growing tissues between them and the ectoderm, except a few small muscles and connective tissue, they are not pressed or shoved out of the way very much, and are so able to retain their original position. I am inclined, however, to the former view, that there is a definite functional relation between these glandular cells and the tracheæ. That the \l=ce\nocytes in Zeuzera pyrina are so large is probably due to the prolonged larval stage of this insect. As previously stated, it remains a larva three years, during which time it eats ravenously, and grows very heavy, acquiring an enormous amount of fat. Naturally, since it eats so much, and stores up so much reserve food, it has a great deal of material to oxidize, and consequently needs a large supply of oxidizing enzymes. A comparison with other insects would lead me to this view, for those having a short larval stage like the Dipteran larvæ, have much smaller \l=ce\nocytes in comparison with the remainder of the body than forms with a prolonged larval life, like the phryganeid larvæ, for example. Before closing, I wish to express my thanks to Professor William M. Wheeler for the kindly advice and encouragement which he has given me at all times. I also wish to thank my father, Dr. Charles Glaser, of Baltimore, for many valuable technical points.