responds to such a change of conditions may be seen from the experiments of M. Le Sage. It is usual to attribute. the fleshy character of the leaves of several maritime plants to their growth by the seaside where salt can reach them; and that salt is the cause of fleshiness is believed from a large number of coincidences, coupled with the fact that the same plants are not always fleshy when growing inland. This hypothesis was confirmed by experiment. M. Le Sage succeeded in producing the characteristic fleshiness of maritime plants in inland plants by watering them with salt water; and the alteration of the tissues was carried on in the second generation from the point gained in the first.*

Another change, of which we have evidence at the present time, and which doubtless has taken place during the geological ages, is the gradual drying up of lakes.

"At the north end of Lake Nyasa, old natives will point to a ridge of sand ten or more feet in height above the present level of the lake, and tell you that they remember the water being at that point; while there is undisputed evidence from observations to show that the present average level of the lake is several feet below what it was ten or twelve years ago. Lake Shirwa, which may be looked upon as

neither more nor less than a vast flat basin, has receded on the western shore at least a mile, within the memory of lads under twenty-five years of age."-(The Geographical Journal. vol. i., p. 247.)

The effect of such a change of external condition may be seen from certain observations and experiments made by Karl Semper, who accidentally observed that the common pond-snail (Lymnaea stagnalis) was so remarkably sensitive to the effects of the volume of the water, that in the space of six days the difference in the length of those living in different volumes of water could be easily and

* Henslow. Natural Science. vol. i., p. 176.

accurately determined. In order to demonstrate that this change was really due to the volume of water, he instituted a series of experiments, in which he took care that all the other conditions of healthy existence should be the same. In all his experiments the only difference was the volume of water, with this result:-The smaller the volume of water which fell to the share of each animal, the shorter its shell remained.

"The first of the shells formed in 100 cubic centimetres of water attained a length of only six millimetres; the second, in 250 cubic centimetres, was nine millimetres long; the third, in 600 cubic centimetres, was twelve millimetres; finally, the fourth grew to eighteen millimetres in 2,000 cubic centimetres of water. . . . These animals, with such immense differences in length, were all the offspring of one mass of eggs simultaneously transferred, and had all reached the same age of sixty-five days."—(p. 162.)

Dr. Henry de Varigny admits the fact, although he offers a different explanation to that which Semper suggests.* The fact is all that my present argument requires.

But the growth of the pond-snail also depends upon temperature. The pond-snail cannot grow if the temperature is below 12° centigrade. Hence an unusually cold season would prevent the natural growth of the animal.

It should next be observed that this animal grows for at least a half of its life.

"This mollusc," says Semper, " grows at a very moderate rate; individuals brought up even under favourable circumstances, take about three months to develope a shell twenty-four millimetres long, and they do not attain their full size under two years, although the whole life of the individual can scarcely exceed three or four years at the utmost.”—(Animal Life. p. 108.)

And not only is there ample opportunity for the low temperature to take effect in dwarfing a particular genera

tion, but it is also quite clear that this dwarfed condition will be inherited.

"Assuming that a young Lymnaea were placed in a lake or stream, of which the temperature constantly exceeds the minimum at which the snail can begin to grow, during only two months of the year, while it never perhaps reaches the high optimum 25°, the mollusc will be unable to attain its due proportions during the first year, or to grow to its full size even during the second, and thus a dwarfed form will inevitably arise. This dwarfed form will still be able to reproduce and multiply itself; for the maturation of germinal matter-the ovum and sperm--takes place during the winter and early spring, at a time when the low temperature of the water hinders all growth, and the optimum of warmth for the sexual processes is much lower than that for growth. Thus a permanently diminutive race might arise if the conditions of temperature above described remained constant for several successive years in the lake or stream where the young molluscs or the eggs have been deposited. Hence it has been supposed, and in many cases no doubt with justice, that the dwarfed races of animals which are found on high mountains or in the polar regions-where they must meet with the conditions of temperature just describedhave originated directly from the low temperature hindering their growth. This assumption, as is quite evident, perfectly accords with my experiments on Lymnaea."-(Semper. Animal Life. pp. 108-9.)

It is obvious that any cause which retarded the growth of the Lymnaea, until the period of sexual maturity, would produce a similar result. In this, as in previous instances, we have examples of the direct action of changed conditions, apart from the principle of selection.

CHAPTER V.

NATURAL SELECTION NOT MANIFESTED IN ORGANIC

EVOLUTION (continued).

(c) EMBRYOLOGY.

"Une échelle des êtres."-BONNET.

THE strongest argument in favour of Organic Evolution, and the one which supplies the key to all the others, is the science of Embryology, which deals with the evolution of the individual. The successive stages through which the organism passes from the fecundated ovum to the mature structure display an increasing complexity. This development of the individual organism is regarded as the "abstract and brief chronicle" of the evolution of the race. The ontogeny or the history of the development of the individual organism, is taken as a proof of the phylogeny or the history of the development of the tribe or race from which the individual has been derived. It is true that there are certain limitations to this principle, but these limitations do not in any way militate against the general inference in favour of Organic Evolution; they simply modify our views as to the precise details of the development which has taken place. Assuming that the development of the individual is a reproduction of the evolution of the race, and that each individual "climbs up its own ancestral tree," we have now to ask whether we can find any trace of the action of Natural Selection in the process

of individual development which would entitle us to infer that it has also been a factor in the evolution of the race.

If all animals were born in a perfect form, except so far as growth and the development of the sexual elements were concerned, we might have some difficulty in drawing any inference on the subject; for it might be contended that the development which took place before birth occurred independently of external conditions, and was the result of heredity; and that it was therefore impossible to investigate the causes which brought about these changes in the first instance. Even if this were so, we might glean some hints upon the subject. In some instances, the young while yet unborn are fully equipped for an aquatic life, although destined when born for a terrestrial one. The Alpine salamander is born in the fully developed condition without gills. In the uterus, it is provided with very long gill-filaments. And the inference from this fact is, that the ancestors of the animal were gill-breathers before they became lung-breathers.

But sometimes the process is very much curtailed, and it might be dangerous to draw our inferences too freely. But fortunately for our present enquiry, many animals are not born with the perfect forms which they subsequently acquire, and we are thus enabled to make these imperfect or larval forms the subject of observation and experiment; and the result of this investigation favours the idea that nature has exercised a transforming, as opposed to a selective, influence. It is interesting, at this point, to observe that the difference in the stage of development at the time of birth is determined by circumstances.

"The ringed snake lays eggs which require three weeks' time to develope. But when it is kept in captivity, and no sand is strewn in

*Balfour. vol. ii., p. 142.