254 



SCIENCE. 



[Vol. XL No. 277 



separating species, for many cases were found where species were 

 fertile when crossed. The broad fact remains, however, that, in 

 spite of many exceptions, the rule is that different species, when 

 crossed, do not produce fertile offspring; and I do not think this 

 conclusion is doubted by any one. Though the difficulty is les- 

 sened by the experiments on cross-breeding, it is not removed ; but 

 the difficulty does not lie exactly where it is usually put. The dif- 

 ficulty is not that species are sterile when crossed, but that varieties, 

 however diverse they may be, are always fertile. It is not difficult 

 to understand why the descendants from a common form, should, 

 by the principle of divergence of character, become so unlike each 

 ■other as to be incompatible with each other when crossed. The 

 difficulty lies rather in the fact that in all the experiments of breed- 

 •ers there has been no approach toward the production of sterility 

 between the varieties produced. Breeders have succeeded in pro- 

 foundly modifying animals, and in producing a great number of 

 ■diverse varieties. Sometimes these varieties show greater differ- 

 ences than are shown by separate genera or families of wild ani- 

 mals. And yet there is no tendency observable toward the pro- 

 duction of sterility among these varieties, perfect fertility being the 

 •universal rule. To explain why a large amount of structural dif- 

 ference in domestic varieties should be accompanied by complete 

 fertility, while in a state of nature very slight differences should be 

 attended by sterility, in many cases at least, is to my mind the only 

 difficulty arising in connection with the sterility of hybrids. 



As an explanation of these facts, it has been pointed out that 

 ■domestication has a direct effect upon the reproductive powers of 

 animals, sometimes producing sterility, and sometimes increased 

 fertility. This factor has been suggested, therefore, as explaining 

 ■why the varieties of domestic animals have not become infertile. 

 But the differences to be explained are very great. Most excellently 

 was this matter illustrated by Professor Clark at the last meeting 

 •of the American Society of Naturalists. For illustration he used a 

 large number of mounted specimens of pigeons obtained from dif- 

 ferent fanciers, and a series of mounted sparrows which may be 

 found everywhere. Among the pigeons the greatest profusion of 

 ■color, size, shape, length of bill, etc., was observable, all within the 

 limits of the same species ; while among the sparrows a sharp eye 

 was required to see any differences between species, and some- 

 times between genera. Allowing what we will for the effect of 

 ■domestication, it is a remarkable thing that the fantail and powter 

 will breed together perfectly well, so that care must be taken by 

 ■the breeder to keep them separate ; while the different species of 

 sparrows with such close resemblance do remain perfectly distinct. 

 'Of course, also, the existence of varieties in nature cannot be due 

 to domestication. All of these facts seem to indicate that some 

 different process has been at work in the production of species 

 from that which has given rise to these very diverse varieties. 



Now, all of this class of facts receives a ready and natural ex- 

 planation in the hypothesis suggested above. All domestic varie- 

 ties have been artificially preserved by man, and he has naturally 

 -selected for preservation such peculiarities as are particularly pleas- 

 'ing or useful to him. It is plain enough that he has not included 

 in his selection peculiarities of the sexual organs : for these are 

 frequently not visible, and have never been the object of improve- 

 4nent on the part of the breeder. Plumage color, shape, size, 

 strength, swiftness, etc., have all received attention ; but I have 

 yet to hear of a single instance where sexual variations have been 

 selected. Certainly this has not been done in the pigeons, or dogs, 

 •or other animals, where such great diversity has been found com- 

 patible with perfect fertility. There can be no doubt that the 

 .sexual nature is just as truly subject to variation as any other part 

 of the body. Every one knows of variations in fertility, in size and 

 shape of sexual organs, in sexual passions, all of which plainly 

 indicate, that, though not so evident to observation, variations in 

 the sexual system are as abundant as elsewhere. Further, it is 

 evident that sterility of species when crossed must be due to some 

 •differences in the sexual organs or sexual elements which prevents 

 proper fertilization or proper growth after fertilization. Is it not, 

 then, a natural conclusion that an accumulation of sexual varia- 

 tions will result in sterility, while any accumulation of other varia- 

 tions will not necessarily have the same effect unless they are also 

 accompanied by sexual variations .' Under artificial breeding there 



have been produced anatomical varieties based upon structures which 

 have had no necessary connection with the sexual nature, and 

 hence the varieties have not become sterile. On the contrary, the 

 uniform conditions of experiment, the rejection by the breeder of 

 individuals which have shown abnormal sexual instincts, have 

 tended to prevent the development of any sexual differences suffi- 

 cient to produce sterility. 



Under nature, however, the conditions have been verj' different. 

 There has been no rigid conforming of selections to anatomical 

 differences. Hardships, famines, surplus of food, etc., have all 

 had their effect ; and there is no part of the body so soon affected 

 by such changes as the reproductive system. Animals have had 

 every opportunity for the free exercise of every passion, and thus 

 differences in the reproductive system have come in for their share 

 in accumulation by natural selection, or otherwise. Romanes is 

 indeed inclined to think that such variations will be specially favor- 

 able for preservation, since they will tend to prevent crossing of 

 unlike individuals. This is, however, doubtful ; but it is plain 

 enough that they will have a much more favorable chance for 

 preservation than they do have under domestication. By variation 

 in this direction there may thus be produced species which will be 

 sterile when crossed, and yet with very small anatomical differ- 

 ences. On the other hand, there may be varieties which would 

 differ widely in anatomical characteristics, and yet be perfectly fer- 

 tile when crossed. The difference between a highly variable species 

 and constant species would be thus due to the readiness with which 

 variations in the reproductive system are produced and preserved. 

 Where the reproductive system is constant, there may arise a 

 highly variable species ; but where the reproductive system is 

 highly variable, there will be a tendency to the production of 

 numerous closely allied species. All of this will lead to a new 

 understanding of the significance of species as groups of animals 

 in which variations have largely affected the sexual organs, with 

 sometimes great and sometimes little change in other parts of the 

 body. In varieties, on the other hand, variation may have affected 

 any other part of the body to almost any degree, but has not af- 

 fected the sexual system. This understanding is somewhat differ- 

 ent from that of Darwin, since it does not regard a species simply 

 as an exaggerated variety. Sometimes it may be so, since anatom- 

 ical and sexual variations may accompany each other. Sometirries, 

 however, a species may be produced directly by sexual variation, 

 without passing through any prominent stage, in which it is a 

 simple variety. Variety and species are therefore independent, 

 being founded on different kinds of variation. 



A discussion of this hypothesis is not possible here, the design of 

 this note being simply to call the attention of American naturaUsts , 

 anew to the subject, and to state the hypothesis as it lies in the 

 mind of the writer. It would be a very important series of experi- 

 ments if some one who has opportunities for experimental breed- 

 ing would undertake the production of a distinct species by select- 

 ing sexual rather than anatomical variations. Such a series of 

 experiments might solve the question of the origin of species. 



H. W. Conn. 



Middletoivn, Conn., May 15. 



The Ohio Mounds. 



The evidence brought to light by the explorations of the Bureau 

 of Ethnology bearing upon the authorship of the typical ancient 

 works of Ohio, leaves scarcely a doubt that these structures are to 

 be attributed to the Cherokees. The chain connecting the Chero- 

 kees of modern times with the builders of these celebrated works 

 seems to be so complete as to leave no break in which to thrust a 

 doubt. 



These explorations have also thrown some light on the so-called 

 clay ' altars ' of the Ohio mounds, rendering it probable that they 

 were places for torturing prisoners of war, — the chief ' sacrifices ' 

 the Indians were in the habit of making. Strange as it may seem, 

 the chain of evidence on this point reaches into actual history. 



The ancient works of Ohio are attributable to at least four, but 

 probably more tribes. Cyrus Tho.mas. 



Youngsville, Penn., May 14. 



