502 Messrs. Farmer, Moore, and Walker. On the Cells [Dec. 8, 



as a pathological one, did it not form a normal stage of the life- 

 history of the species. 



We have said that the cells from which the heterotypically dividing 

 elements will finally arise can often be distinguished from those cells 

 which will not produce such elements. In the testis of a mammal or 

 in the sporogenous tissue of a stamen we recognise with ease and 

 certainty the existence of these cells. They continue to multiply, and 

 though differing from the adjacent cells in many respects, they continue 

 to resemble them in their mode of nuclear division until they pass 

 severally into the peculiar state of growth that ushers in the hetero- 

 type division. 



In our studies of abnormal growths occurring on ferns, we were 

 struck by certain features presented by the proliferating tissues that 

 are formed during apogamy and apospory, and we have thus been led 

 to make a systematic investigation of the cytological features presented 

 by malignant growths in man. 



This has resulted in the recognition of the existence of a surprising 

 degree of similarity between the phases that characteristically recur in 

 such tissues and those transformations of somatic cells into reproductive 

 tissues in general. 



Thus in a typical example of rapidly growing epithelioma it is seen 

 that in the early stages of the proliferation of the Malpighian layer, 

 the cells of the invading tissue at first pass through a cycle of somatic 

 divisions, exactly as in the early stages of reproductive tissue. The 

 resemblance may extend to the frequent production of giant cells, a 

 common occurrence in each case. 



As cell multiplication proceeds, however, a change passes over the 

 cells themselves. The protoplasmic continuity, to which the " prickly " 

 character is due, becomes more or less obliterated, and the cells assume 

 that appearance of indifferent germ tissue so well known as a feature of 

 the elements of which malignant growths are largely made up. But, in 

 addition to this, other important changes occur which seem to have 

 been generally overlooked. 



A varying number of cells, situated in a zone behind the growing 

 edge of the advancing neoplasm, may be observed to attain some- 

 what large dimensions. Each contains a nucleus that grows to a 

 considerable size. As the latter enters on a prophase of division, 

 it is recognised that the chromosomes, instead of appearing as 

 delicate thin rods or V's, which are split longitudinally, present the 

 appearance of short thickened loops or rings, closely resembling the 

 later prophase stages of the heterotype mitosis in the normal repro- 

 ductive tissues. What is still more significant is the fact that in these 

 cells the number of the chromosomes is obviously less than in the normal 

 somatic cells of the surrounding tissues. In many cases we determined the 

 numbers to be approximately halved as compared with those of the 



