those colors that have the greatest illuminating power are those that least favor germination.

Ad. Brongniart (1832) announced as the results of his experiments that the retarding influence of light depends not only on the illuminating power of the colored light, but on the relative quantity of white light that passes through the different colored glasses. In all these experiments the seeds were several millimeters below the surface of the soil, so that the colored lights did not affect the seeds directly, but indirectly through the soil whose temperature and moisture and evaporation may easily be of predominating importance.

Ph. A. Pieper (1834), Meyen (1837), Zantedeschi (1846), and Belhomme (1854) have all experimented on the growth of seeds under colored glasses; but the sources of error incident to this method of observation prevent us from drawing any conclusion as to the influence of light itself.

Ville (1865) says that the injurious effect of solar radiation on germination is the result of the heat only and that the effect of the light is inappreciable. For aquatic plants whose seeds germinate in the water, darkness seems decidedly favorable to germination, but it acts only in an indirect manner by preventing the warming of the water and the disengagement of the oxygen that is dissolved in this

water.

Charles Darwin (1877) says that certain species of seeds do not grow well when they are exposed to the light, even the diffuse light of

a room.

Duchartre (1877) considers the action of darkness as a secondary influence, useful but not at all essential and concerning which there has been too much exaggeration.

Faivre (1879) has shown that the appearance of the primordial latex occurs at a moment when the radicle is only a few millimeters long and when the cotyledons are still inclosed in the seed envelopes and have not yet received the action of light. He notes that under a yellow light obtained by transmitting sunlight through a solution of bichromate of potash the seeds develop their chlorophyl and their latex more rapidly, and consequently have a shorter period of germination than under a blue light obtained by transmitting sunlight through a solution of the ammoniacal oxide of copper.

Detmer (1880) has consecrated an extensive work to the study of the germination of seeds, and states that concerning the action of light we are still ignorant as to whether it is direct-that is to say, whether it stimulates the storing up of new substances in the vegetable tissue or whether, on the contrary, it strengthens the persistence within the cells of some special process having a more or less intimate relation to the phenomena of growth and which can only

proceed in darkness. Detmer adds a few historical references, viz, Humboldt (1794), according to whom seeds sprout more easily in darkness than in light; Fleischer (1851), Heiden (1859), and Nobbe (who all consider solar rays as having no action on the seeds), and, finally, Hunt (1851), who considers that light retards germination.

After this preliminary historical survey, Pauchon communicates the results of his own experiments as to the influence of light on germination on the following twenty-two species of plants:

After deducting doubtful results or failures Pauchon gives the following conclusions (see p. 131 of his work above quoted):

(1) In 22 experiments germination occurred first in the light; in 26 experiments it occurred first in the dark.

(2) Five times we obtained duplicate results favorable to the light for the same species of plants (Arachis, Zea mays, Dolichos, Sinapis, and Linum). Eight times these duplicate results were favorable to specimens kept in the dark (Helianthus, Delphinium, Pancratium, Ricinus, and Papaver). In one case (Linum) two results were obtained favoring light and two favoring darkness.

(3) Among the 22 species of plants used in the experiments 14 gave mixed results equally favorable whether placed in the light or the dark.

(4) Among the 8 other varieties only 1 gave negative results (Coffea); 3 gave results favorable to light (Cucurbita, Spilanthes, and Carthamus); 4 gave results favorable to darkness (Delphinium, Pancratium, Lepidium, and Nigella).

It appeared to Pauchon impossible to draw any conclusion whatever from these facts. Should we be astonished at this? The problem is certainly much more complex than appears at first sight.

There is every reason to suppose, for example, that the action of light is not the same under all the conditions of temperature which obtained during these experiments. Here again, however, we are confronted by the unknown; because, in order to draw from these researches the consequences which might flow from them it would be necessary to know precisely the thermic conditions favorable to the germination of each species. Unfortunately this is a very important gap to be still filled up, as the work accomplished in this direction gives only approximate results limited to a very small number of different kinds of seeds. On the other hand, looking to facts of another order, mentioned further on in this work, we think that we may be allowed to suppose that the influence of light can only be favorable to germination when it acts at temperatures below that which is most favorable to germination. A considerable number of observations already cited would seem to be in accord with this view of the subject. But unfortunately the many contradictions that we observed in our results do not allow us to accept this opinion as based upon a solid foundation.

Pauchon then goes on as follows:

Another reason, however, induces me to admit, only with many reserves, the results of experiments whose critical epoch is the visible development of the embryo. A method based on this special observation does not appear to me capable of furnishing a really scientific basis for the determination of the question before us. The process of germination is not, in reality, as simple a phenomenon as the greater number of botanists, perhaps too easily, take for granted. Its complexity is even so great that one can not judge of the actual development of the germ of the plant and of the degree of its physiological activity by the external characters observable by the eye, such as the bursting of the spermoderm and the more or less rapid protrusion of the radicle. I do not hesitate to say, according to observations frequently repeated, that this is an empirical process and entirely deceptive in the particular case that we are dealing with. Although it may be capable of furnishing valuable results when we wish to judge of the influence of some one of the fundamental conditions of germination, it becomes utterly insufficient when it is a question of observing the more delicate and fugitive influences, such as that of light. I have, in fact, in the course of chemical researches, given in the next chapter, demonstrated that for the same stage of apparent development the absorption of oxygen by the seeds in the process of germination varies to a large extent with the temperature, and has no relation to the external growth of the embryo. It is, however, not surprising that the development of the embryo continues in the interior of the seed for a much longer time in one seed than in another of identical appearance; the unknown and variable relation between the reserved nutrition and the rudimentary vegetable is probably the explanation of these hitherto unexplained peculiarities. Although the researches given in this chapter do not give any positive result on the subject of my work, I have preserved them and pub

lish them here in order to explain to observers the defects of au experimental process to which, in the future, they would themselves have been tempted to resort; this, moreover, seems to me the more useful in that up to this time this danger does not seem to have struck the attention of botanists. On the other hand, my observations contain some new data relative to the temperatures favorable for the germination of certain exotic seeds.

In consequence of the conclusions to which we have thus been led, it would be useless to study the action of the different portions of the solar spectrum on the apparent progress of germination. How, in fact, can we suppose, in view of the contradictory results already obtained for the condition of light and of darkness-that is to say, for the most extreme conditions-that the employment of the same method can reveal a difference of action for the various portions of the spectrum?

Is it then necessary, after this first fruitless attempt, to give up the solution of the problem, or shall we seek it by another and better method? It is this latter alternative that I have adopted in that I have taken for the basis of a new series of observtions the variations of a physiological process that, in an almost mathematical manner, measures the germinal activity of the vegetable embryo, namely, the respiration.

After giving the details of his experiments on respiration of plants, Pauchon draws the following conclusions (p. 166):

The laws brought prominently forward by the results of these experiments are:

(1) Light exercises a constant and more or less marked accelerating influence upon the absorption of oxygen by seeds in the process of germination. All the experiments made in a strong light have not, however, the same value in demonstrating this fact. But if we have doubts about the precision of the results furnished by experiments in which germination did not invariably take place (and we believe that we have shown by some preparatory experiments that these results have at least a relative value), this certainly is not the case with experiments Nos. 2 and 8, in which all the seeds did germinate. Thus experiment No. 2 showed in favor of light a result as to the oxygen absorbed twice as great as that given by the seeds placed in the dark. In the same way in experiment No. 8 this superiority reaches to one-third of the quantity of oxygen absorbed by the seeds placed in the dark. Finally, the other experiments, and particularly those classed under Nos. 3, 6, and 7, further confirm the generality of this action of light, which we will, besides, find again in a second series of experiments reported hereafter, several of which have shown unanimity of germination in both cases.

(2) There exists a relation between the degree of light and the quantity of oxygen absorbed. Thus, in a diffuse light this accelerating influence shows itself in a most marked manner when the sky is very clear, and the solar radiation reaches us in its greatest intensity. Such was the case in experiments Nos. 2 and 8. Whenever the sky is cloudy this action is more and more weakened and ceases altogether when the sun is completely veiled, as in stormy weather, so that there is a semiobscurity.

However, in all the experiments where the final result has been favorable to the action of light I have convinced myself that a cloudy sky for twelve hours always showed itself in the amount of the absorption of oxygen in such a manner that the examination of these figures, noted day by day, would almost serve to show the state of the atmosphere during the day which preceded the observation. A very conclusive instance of this action is given us by experiment No. 4 of the second series, in which the state of the sky being carefully observed it showed very marked changes.

(3) The accelerating influence exercised upon seeds exposed to the action of light during the day did not stop at night; it continued to act in the dark with an equal, sometimes even with a greater intensity. I will cite as examples experiments Nos. 3, 4, 6, 7, and 8, when observations made twice a day, morning and evening, allowed of examining the fact I state. How can we explain this persistent action of light? One hypothesis only can be admitted. A portion of the action of the light absorbed by the grain during the day is stored up by it and used by it at night to accelerate its respiration. The proof of this is that the differences of elevation [or quantities of absorbed oxygen] shown in the morning by the instruments for seeds kept in the dark are always below those shown by the instruments and plants in the light. The influence of the light, then, continues for a certain time, at least several hours, even after the light itself has ceased to act; on the other hand, however, this action is not exerted immediately. There is one other phenomenon that we have demonstrated by our experiments. Suppose the sky to be very clear; the differences in favor of light are only apparent after two or three days and become much more marked toward the end of the experiment; that is to say, in proportion as the daily action of sunlight is more and more frequently repeated.

(4) I should also call attention to still another peculiarity, viz, that the differences in the quantities of oxygen absorbed in the dark and in the light were generally much greater at the beginning of these researches than in the later experiments, and particularly in those of the second series. The temperature appears to me to be the only element that varied in these experiments. There must therefore be a more intense respiratory action exercised by light at low temperatures, and this influence would become weakened at high temperatures. This fact would be in entire agreement with the demands of physiology. It is easy of comprehension that a scarcity of heat should be counterbalanced by the action of light, which furnishes for the reaction of the respiratory organs the force that they could not obtain from an insufficient temperature. On the contrary, when the heat is intense the intervention of the light is no longer necessary, the first cause being sufficient to excite the process of germination in the protoplasm of the seeds.

(5) This action of light seems to differ a little according as it acts upon seeds containing albumen or those without albumen. case of the albuminous seeds of the castor-oil plant the advantage was much more apparent in favor of those exposed to the light, which advantage appeared to me much less decided for the seeds without albumen, such as the haricot bean. Nevertheless, as the experiments were not invariable in their results, the cause of the variations ob