October 24, 1884.] 



SCIENCE. 



399 



how far sexual reproduction extends down- 

 ward in the scale of life. I deem it very 

 probable that it extends to the lowest animat- 

 ed being, even though it be quite differently 

 manifested in the lower forms from what we 

 observe in ordinary bi-sexual reproduction. 

 This view is opposed to the opinions generally 

 held : for botanists trace the evolution of sex 

 within the vegetable kingdom ; and zoologists 

 trace it, though less definitely, within the ani- 

 mal kingdom. We are thus forced to assume 

 that sex, one of the most fundamental and 

 characteristic phenomena of life, has arisen 

 twice. This is to the last degree improbable. 

 Such a coincidence would be the most extraor- 

 dinary result of chance within human experi- 

 ence. It is more reasonable to suppose, that, 

 though we do not yet recognize it, the sexual 

 function exists in the protobionts, which are 

 neither animal nor vegetable, and that they 

 also produce a body homologous with an im- 

 pregnated ovum ; and to suppose, further, that, 

 out of this common commencement, both ani- 

 mal and vegetable sex have been evolved. The 

 essential property of the sexually produced 

 ovum is its power of repeated division, pro- 

 ducing a succession of cell-generations, which, 

 together with the original body (ovum), con- 

 stitute the cycle. There is much evidence of 

 a positive character to confirm the belief of 

 the cyclical course of life, even among the pro- 

 tozoa and protoplrytes, in which there occurs 

 what is known as rejuvenation (verjiingmig) . 



2°. I maintain that it is probable that all 

 cycles of cells are self- limited. Let us first 

 ascertain the nature of the limitation. Our 

 knowledge of the manner in which the cycles 

 are limited (i.e., of the causes of natural 

 death) is very restricted, and derived solely 

 from the higher animals. My own special in- 

 vestigations have been in this field, and have 

 led me to the opinions and problems we are 

 discussing. 



M3' experiments demonstrate, that, when 

 properly analyzed, the growth of at least the 

 higher animals gradually diminishes from birth 

 onwards, almost without interruption. This 

 is an irrefutable mathematical verification of 

 the views which I advanced in nry article on 

 'Growth as a function of cells,' published in 

 1879, the essence of which, as far as we are 

 now concerned, is, that the cells of a cycle 

 continuously lose their power of division, so 

 that the interval between two successive divis- 

 ions gradually increases. This involves the 

 ultimate termination of the cycle, because the 

 losses go on, not only until the cells can no 

 longer divide, but until the} 7 exhaust them- 



selves. This whole series of changes is prop- 

 erly senescence, or growing old. Senescence is 

 a continuous process, covering the whole period 

 of a cycle of cells ; and we must assume it is 

 the positive loss of power in the single cells, 

 such that the last-produced cells cannot con- 

 tinue, and natural death ensues. Of course, in 

 the cases of a multicellular animal, death of 

 the whole follows secondarily upon exhaustion 

 of any essential part ; as in the case of insects, 

 which die upon laying their eggs. In the 

 higher animals, then, the Cj'cle is limited b} T 

 senescence, and senescence is a decay which 

 probably begins when the cycle begins. The 

 next point to decide is, whether the same phe- 

 nomenon occurs with the unicellular organisms. 

 If it is found that the divisions of a Parame- 

 cium, 1 for instance, after a conjugation, are at 

 first rapid, and then follow at increasing inter- 

 vals, it would prove (provided, always, the ex- 

 ternal conditions remained constant) that we 

 here had true senescence, with its sequel, natu- 

 ral death, or the end of the cycle. Until this 

 point is settled, we cannot know whether there 

 is, among unicellular animals, a form of death 

 homologous with the natural death from senes- 

 cence in the higher animals and plants. 



It is to be regretted that both Weismann 

 and Goette appear not to know the article to 

 which reference has just been made : otherwise 

 they would have recognized that the problem 

 of death is, first, whether growing old (veral- 

 tung, involution) is a universal phenomenon of 

 life. Weismann' s first article was an address 

 delivered before the German naturforscherver- 

 sammlung, September, 1881, and subsequently 

 republished at Jena. 2 He advanced then the 

 view, that, for unicellular organisms, there is no 

 death except through accident ; that, the propa- 

 gation being by simple division, we must as- 

 sume that the process of division ma}* go on 

 forever. He does not even consider whether 

 the cells form cycles, and whether these c}'cles 

 need to be renewed ; so that he misses the real 

 problem. On the contrary, he is enchained a 

 prisoner to the mystical idea of individuality, 

 and reasons as if individuality rendered direct 

 comparisons legitimate between things essen- 

 tially different. All his reasoning is based 

 upon the idea that an individual protozoan is 

 comparable to an individual dog, and so on. 

 The argument just made against him was to 

 show that the basis of his whole fabric is illu- 

 sory. Butschli, in his short article, 1 called forth 



1 Paramecium is a common unicellular animal. 



2 Weismann, Ueber die dauer des lebens (Jena, 1882, 8°), 

 94 p. Cf. also Weisraann's comments on Butschli, Zool. anzei- 

 ger, v. 377-380, and his reply to Goette, — Ueber leben und 

 tod (Jena, 1884, 8°). 



