RELATIVE VIABILITY IX MAMMALS AxVD BIRDS 541 



longevity *, the limit of age which an individual could attain in 

 the most favourable circumstances, and aveiage or specific 

 longevity, the average age to which the animals of a species 

 attain under the natural conditions to which the species has 

 become adapted. The diflerence between these two, I suggest is 

 a measure of the severity of the conditions to which the species 

 is subjected. In the case of the vast majority of animals it is 

 impossible to get information as to either average or potential 

 longevity. What information we have, is derived chiefly from 

 knowledge of animals in captivity, and in this memoir I have been 

 able to set down further information about species which have 

 already been discussed by J. H. Gurney, Metchnikoff, IBrehin and 

 others, and to add information regarding many species about 

 which nothing has been recorded hitherto. 



(3) I apply the principle of comparison between average and 

 j)otential longevities to the case of animals in captivity, by 

 discussing the meaning of average duration and maximum 

 duration. I suggest that a measure of the efiect of the conditions 

 to which animals are subjected in captivity may be obtained by 

 comparison of the maximum duration with what may be known 

 from other sources of the potential longevity, and by comparison 

 of the average duration with the maximum duration. If tlie 

 individual facts were collected in such a fashion as I have 

 suggested in paragraph 1 of this general summary, I think the 

 new method would lead to incontrovertible results, but even with 

 the facts at my disposal, conclusions of fair validity can be drawn. 

 It is necessary to note, however, that the figures of individual 

 duration could be used to reveal more, if they were plotted out in 

 curves instead of being used to give an arithmetical mean. The 

 grouping of the individual cases with regard to the mean is of 

 great practical importance. In some cases, for instance, I have 

 pointed out that the majority of the cases w'ere grouped towards 

 the two extremes, that the individuals for the most part either 



* [Lankestev (i^. c. p. 27) pointed out that some orsritnisins, such as fisli, molluscs, 

 larjre crustacea, sea-covals, and many trees, appear to have no tixed potential lon<;'evity 

 but to persist until they perish from disease, or are overwhelmed by some accident, 

 whilst others, such as man, appear to havea set period to the possible duration of their 

 lives. The mammals and birds with which I am concerned in this memoir heloiig: 

 to the second categor\-. But I doubt if there be a fundamental distinction between 

 the categories. First, as Lankester stated, the individuality of an organism like a tree, 

 or a colonial invertebrate, differs from that of most organisms, and must be separately 

 considered from the point of view of potential longevity. Next, certain organisms, 

 such as many fish, appear to have more indefinite limits of growth (possibly related 

 to mechanical factors, of similar importance in the case of aquatic mammals such 

 as the Cetacea) than the majority of animals, the adult dimensions of which vary 

 within narrow limits ; definite or indefinite potential longevity may be related with 

 definite or indefinite growth. (See also H. Spencer, 'Principles of Biology,' revised 

 edition, 1898, vol. i. p. 135). I ant more inclined to suppose that potential longevity 

 varies with structure in the widest sense, including under the term structure the 

 consequences of wear and tear, and auto-poisoning from the slow accumulation of 

 waste-prodttcts. All ])ieces of machinery, from a motor-car to a cork-screw, have 

 a potential longevity, dependent in the main on their structttre, and apart from tlie 

 quality of their manufacture, varying fairly closely with their complexity. Altliough 

 our attention is arrested by extreme cases, these are only terms in a series.] 



