September ii, 1890] 



NATURE 



475 



esculent a, i.e. in which a complete aortic arch was present before 

 the gills were formed. 



We are therefore justified in concluding that as regards the 

 development of the branchial blood-vessels, R. esctilenta has 

 retained a primitive ancestral character which is lost in R. tern- 

 poraria, and it is interesting to note that were our knowledge of 

 the development of amphibians confined to the common frog, 

 the most likely form to be studied, we should, in all probability, 

 have been led to wrong conclusions concerning the ancestral 

 condition of the blood-vessels in a point of considerable im- 

 portance. 



A matter which at present is attracting much attention is the 

 question of degeneration. 



Natural selection, though consistent with and capable of 

 leading to steady upward progress and improvement, by no 

 means involves such prrigre<;s as a necessary consequence. All 

 it says is that those animals will, in each generation, have the 

 best chance of survival which are most in harmony with their 

 environment, and such animals will not necessarily be those 

 which are ideally the best or most perfect. 



If you go into a shop to purchase an umbrella, the one you 

 select is by no means necessarily that which most nearly ap- 

 proaches ideal perfection, but the one which best hits off the 

 mean between your idea of what an umbrella should be and the 

 amount of money you are prepared to give for it : the one, in 

 fact, that is on the whole best suited to the circumstances of the 

 case, or the environment for the time being. It might well 

 happen that you had a violent antipathy to a crooked handle, or 

 else were determined to have a catch of a particular kind to 

 secure the ribs, and this might lead to the selection, i.e. the 

 survival, of an article that in other and even in more important 

 respects was manifestly inferior to the average. 



So it is also with animals : the survival of a form that is 

 ideally inferior is very possible. To animals living in profound 

 darkness the possession of eyes is of no advantage, and forms 

 devoid of eyes would not merely lose nothing thereby, but would 

 actually gain, inasmuch as they would escape the dangers that 

 might arise from injury to a delicate and complicated organ. In 

 extreme cases, as in animals leading a parasitic existence, the 

 conditions of life may be such as to render locomotor, digestive, 

 sensorv, and other organs entirely useless ; and in such cases 

 those forms will be best in harmony with their surroundings 

 which avoid the waste of energy resulting from the formation 

 and maintenance of these organs. 



Animals which have in this way fallen from the high estate 

 of their forefathers, which have lost organs or systems which 

 their progenitors possessed, are commonly called degenerate. 

 The principle of degeneration, recognized by Darwin as a possible, 

 and, under certain conditions, a necessary consequence of his 

 theory of natural selection, has been since advocated strongly by 

 Dohrn, and later by Lankester in an evening discourse delivered 

 before the Association at the Sheffield meeting in 1879. Both 

 Dohrn and I.ankester suggested that degeneration occurred 

 much more widely than was generally recognized. 



In animals which are parasitic when adult, but free-swimming 

 in their early stages, as in the case of the Rhizocephala, whose 

 life-historv was so admirably worked out by Fritz Miiller, 

 degeneration is clear enough : so also is it in the case of the 

 solitary Ascidians, in which the larva is a free-swimming animal 

 with a notochord, an elongated tubular nervous system, and 

 sense organs, while the adult is fixed, devoid of the swimming 

 tail, with no notochord, and with a greatly reduced nervous 

 system and aborted sense organs. 



In such cases the animal, when adult, is, as regards the totality 

 of its organization, at a distinctly lower morphological level, is 

 less highly differentiated than it is when young, and during in- 

 dividual development there is actual retrograde development of 

 important systems and organs. 



About such cases there is no doubt ; but we are asked to ex- 

 tend the idea of degeneration much more widely. It is urged 

 that we ought not to demand direct embryological evidence 

 before accepting a group as degenerate. We are reminded of 

 the tendency to abbreviation or to complete omission of ancestral 

 stages of which we have quoted examples above ; and it is 

 suggested that if such larval stages were omitted in all the 

 members of a group we should have no direct evidence of 

 degeneration in a group that might really be in an extremely 

 degenerate condition. 



Supposing, for instance, the free larval stages of the solitary 

 Ascidians were suppressed, say through the acquisition of food 

 yolk, then it is urged that the degenerate condition of the group 

 might easily escape detection. The supposition is by no means 

 extravagant ; food yolk varies greatly in amount in allied animals, 

 and cases lite Hylodes, or amongst Ascidians Pyrosoma, show 

 how readily a mere increase in the amount of food yolk in the 

 egg may lead to the omission of important ancestral stages. 



The question then arises whether it is not possible, or even 

 probable, that animals which now show no indication of degenera- 

 tion in their development are in reality highly degenerate, and 

 whether it is not legitimate to suppose such degeneration to have 

 occurred in the case of animals whose affinities are obscure or 

 difficult to determine. 



It is more especially with regard to the lower vertebrates that 

 this argument has been employed ; and at the present day, 

 zoologists of authority, relying on it, do not hesitate to speak of 

 such forms as Amphioxus and the Cyclostomes as degenerate 

 animals, as wolves in sheep's clothing, animals whose simplicity 

 is acquired and deceptive rather than real and ancestral. 



I cannot but think that cases such as these should be regarded 

 with some jealou'sy ; there is at present a tendency to invoke 

 degeneration rather freely as a talisman to extricate us from 

 morphological difficulties ; and an inclination to accept such 

 suggestions, at any rate provisionally, without requiring satis- 

 factory evidence in their support. 



Degeneration of which there is direct embryological evidence 

 stands on a very different footing from suspected degeneration, 

 for which no direct evidence is forthcoming ; and in the latter 

 case the burden of proof undoubtedly rests with those who 

 assume its existence. 



The alleged instances among the lower vertebrates must be 

 regarded particularly closely, because in their case thf suggestion 

 of degeneration is admittedly put forward as a means of escape 

 from difficulties arising through theoretical views concerning the 

 relation between vertebrates and invertebrates. 



Amphioxus itself, so far as I can see, shows in its development 

 no sign of degeneration, except possibly with regard to the an- 

 terior gut diverticula, whose ultimate fate is not altogether clear. 

 With regard to the earlier stages of development, concerning 

 which, thanks to the patient investigations of Kowalevsky and 

 Hatschek, our knowledge is precise, there is no animal known 

 to us in which the sequence of events is simpler or more 

 straightforward. Its various organs and systems are formed in 

 what is recognized as a primitive manner ; and the developtnent 

 of each is a steady upward progress towards the adult condition. 

 Food yolk, the great cause of distortion in development, is 

 almost absent, and there is not the slightest indication of the 

 former possession of a larger quantity. Concerning the later 

 stages our knowledge is incomplete, but so much as has been 

 ascertained gives no support to the suggestion of general 

 degeneration. 



Our knowledge of the conditions leading to degeneration is 

 undoubtedly incomplete, but it must be noticed that the condi- 

 tions usually associated with degeneration do not occur. Am- 

 phioxus is not parasitic, is not attached when adult, and shows 

 no evidence of having formerly possessed food yolk in quantity 

 sufficient to have led to the omission of important ancestral 

 stages. Its small size as compared with other vertebrates is one 

 of the very few points that can be referred to as possibly in- 

 dicating degeneration, and will be considered more fully at a 

 later point in my address. 



A consideration of much less importance, but deserving of 

 mention, is that in its mode of life Amphioxus not merely differs 

 as already noticed from those groups of animals which we know 

 to be degenerate, but agrees with some, at any rate, of those 

 which there is reason to regard as primitive or persistent types. 

 Amphioxus, like Balanoglossus, Lingula, Dentalium, and 

 Limulus, is marine, and occurs in shallow water, usually with a 

 sandy bottom, and, like the three smaller of these genera, it 

 lives habitually buried almost completely in the sand, into which 

 it burrows with great rapidity. 



I do not wish to speak dogmatically. I merely wish to pro- 

 test against a too ready assumption of degeneration ; and to 

 repeat that, so far as I can see, Amphioxus has not yet, either in 

 its development, in its structure, or in its habits, been shown to 

 present characters that suggest, still less that prove, the occur- 

 rence in it of general or extensive degeneration. 



In a sense, all the higher animals are degenerate ; that is. 



NO. 1089, VOL. 42] 



