The fifteen tumors which fumished the transplants were typical large invasive adenocarcinomas having the usual irregular and apparently anarchic arrangement of their component tubules and acini. However, the transplanted tumors developed according to definite and well defined structural patterns, their type depending on the immediate physical environment. Three such morphogenetic patterns were observed.

Unimpeded outgrowths into the aqueous tumor characteristically assumed a tubulo-papillary arrangement; the earliest formation consisted of solid, purely epithelial cylinders, many of which at later stages acquired a lumen and thus became tubular; generally only the coarser projections developed vascular stalks. Further growth was made by repeated branching, and lateral outpouching of the tubules. Where the tumor grew in contact with firm, even surfaces, such as lens or cornea, differentiation was lost and broad membranes formed which gradually spread over the surfaces; secondarily, new cylindrical or tubular processes arose from such indifferent cellular carpets. Where the tumor made contact with loose tissue such as iris, it invaded this organ, and, supported by the stroma, assumed an acinar pattern quite like that of the original adenocarcinoma of the kidney. All three types of growth were sometimes found coexisting in different portions of the anterior chamber.

The rate of growth of transplanted tumors was followed by photographic records taken periodically. A fairly constant mode of progress was noted: after a variable period of lag and a period of gradual outgrowth, there followed rather abruptly a short period of rapid growth, after which growth increments gradually became smaller. Variations in growth rate due to season and to temperature were evident, but further experiments are required to evaluate the part played by these factors.

Attempts to transplant normal kidney in the anterior chamber were unsuccessful.

The manner of growth in the vitreus was found to be similar to that in the anterior chamber.

The factors that determine the manner of tumor growth are as yet imperfectly understood. Evidence is however accumulating to the effect that neoplastic growth is not as anarchic as is suggested by histologic sections of some tumors (26). The present experiments support the view that cancers are much more responsive to the laws governing growth and organization than is generally supposed.