438 NATURAL SCIENCE. auoust. 



descend the animal scale. Moreover, in the embryonic stages of the 

 higher animals — in those stages where integration is less complete — 

 fission as an internal or as an external accident produces, not death, 

 but double monsters or identical twins. So, also, if there be con- 

 sidered the suddenness of somatic death. The heart of a higher 

 animal stops suddenly, and this accidental moment is named the 

 moment of death. But the heart of a tortoise that has been in a 

 dissecting dish for two or three days, begins to beat as the knife 

 reaches it. The watchers by the death-bed of a fish or a mollusc 

 would continue to watch until putrefaction was obvious. I have fed 

 a semi-putrid anemone with pieces of mackerel, and have seen a star- 

 fish with the tips of its arms rotten and pulpy crawl gaily about and 

 evert its stomach on a pecten. In both of these cases the animals 

 were moribund, and soon afterwards actually died. It is a common- 

 place of observation how rapidly the lower animals putrefy, and how 

 necessar}', to secure their histological preservation, it is to kill them 

 as soon as possible after capture. 



All this merely means that, in the first place, the stages of death 

 creep over their simple framework so quietly that they are decaying 

 before we know that they are dead ; and, in the second, that proto- 

 plasmic death is more nearly identical with somatic. 



Finally, in the descending scale, we reach unintegrated colonies 

 of cells or single cells. Here there is identity between the proto- 

 plasmic individual and the soma ; there is no somatic mechanism to 

 break down, no chain along which protoplasmic death can creep from 

 one cell to another. 



It is clear, then, that somatic death is an attribute of the 

 metazoa, a merely necessary result of their complexity of organisa- 

 tion, and that somatic death becomes a more and more striking 

 occurrence as the metazoon becomes higher and more complex. It is 

 hardly necessary to point out that a line of argument precisely 

 similar applies to somatic life, and that what are usually meant by 

 life and death are merely somatic life and death, the merest mechanical 

 results of complex mechanism. 



It is true that the protozoa are untouched by somatic death, that 

 it has and can have absolutely no place in their life-cycle ; that were 

 somatic death the only death in the world, protozoa would be not 

 potentially, but actually, immortal. It is equally true that no 

 metazoan, however lucky, is at all likely to escape somatic death, is 

 at all likely to be immortal. Even in the simplest metazoon, the life 

 of the soma is bound up with the life of the units — or, at least, with 

 the life of groups of the units — and the chances of protoplasmic death 

 of any cell, or group of cells, being followed by somatic death, con- 

 tinually increases with the complexity of the organism, and increases 

 in geometrical ratio as the number of component parts liable to 

 accidental or intrinsic break-down. 



The protozoa, then, and chains of single ccU-lifc, like the 



