There seems to be some evidence in favor of the conclusion that the mean molecular weight of the solutes from the abnormal structure is higher than that from the ovary wall. There are 29 cases in which the mean molecular weight for the included mass is higher than that for the ovary wall, to 22 cases in which the reverse is true. In four instances the two values are identical. The averages for the three major series of determinations are as follows:

In all cases the averages for the solutes of the proliferous mass are higher than for those of the fruit wall, but the probable error of the difference is so high that little weight is to be attached to it.

While these results are by no means conclusive they seem to indicate that the solutes in the sap of the proliferous mass have a slightly greater average molecular weight than the solutes in the sap of the ovary wall. Such a conclusion would be in good agreement with the result noted under (D) above, i. e., that the organic materials form a greater proportion of the dissolved substances in the sap of the prolification than they do in the sap of the ovary wall. It is but reasonable to suppose that the organic substances have a greater molecular weight than have the ions and undissociated molecules of the inorganic salts.

5. Summary and conclusions. In Passiflora gracilis prolification of the fruit frequently occurs. This generally consists in the production of a series of whorls of incompletely closed carpels borne on a short stalk arising from the bottom of the fruit cavity, and represents in all probability a continuation of the main axis which gave rise to the carpels entering into the composition of the fruit wall.

As one of the steps in the analysis of the factors to which this highly remarkable teratological variation is due, we have had under investigation for some time the physico-chemical properties of the cell sap of the normal and teratological fruits, and of the parts of the teratological fruit.

In this paper we present data concerning the physico-chemical properties of the cell sap from the tissues of the wall and the abnormal carpellary whorl11 of fruits with trimerous fruit wall and tetramerous prolification. The materials comprise about 60 samples of this class obtained by the dissection of about 100,000 fruits.

The following conditions have been found in the materials examined.

The physico-chemical constants considered-specific gravity, concentration, depression of the freezing point, osmotic pressure, electrical conductance and mean molecular weight-are all susceptible to the influence of the environmental and possibly to the physiological state of the plant upon which they are borne. This is shown by the facts (a) that the mean values of the constants differ sensibly from series to series;12 and (b) that the variability of the determinations within the individual cultures is, in general, larger in the cases in which the collections extended over a longer period of time, and in consequence comprised fruits which had been developed under a wider range of conditions.

The specific gravity of the sap of the fruit wall is of the order 1.0200, while that of the proliferous mass is distinctly lower. Concentration, i. e., weight of solutes / weight of water, is about 0.0400 for the wall and sensibly lower for the included mass. If the differences in specific gravity of the sap from the wall and the included mass be expressed in percentage ratios, with the variable term (d-1) of the specific gravity of the wall as a base, the density of the sap from the mass is, for the averages of the larger series, 5.5, 5.7 and 11.6 percent lower than that of the wall.

The specific electrical conductivity is of the order 0.013000 for the sap of the ovary wall, and about 0.004300 lower for that extracted from the included carpellary whorl. In the three principal series of determinations the conductivity of the sap from the included mass is 29.6, 34.7 and 34.6 percent lower than that from the wall.

The osmotic pressure of the extracted sap, as determined by the depression of the freezing point, is, in round numbers, 7.50 atmospheres in the fluids of the wall and roughly half an atmosphere less in the prolification. For the three large series the mean constants for the abnormal whorl are 12.1, 7.0 and 6.5 percent lower than the values for the fruit wall, which serves as a base for the calculation of the percentages.

The difference between the mean molecular weight of the solutes in the sap of the wall and the abnormal mass, as determined by

is the most variable from sample to sample of any of the constants. Possibly this is due, to some extent, to the fact that an unavoidable experimental error, in any one of three distinct laboratory operations involved in the determination of this constant, would influence the final result. However, the slight absolute variation in mean molecular weight seems to us to indicate a great reliability of the

constants.

The mean molecular weight is of the order 120. There is some evidence that this constant is somewhat higher for the solutes of the abnormal tissues than for those of the fruit wall.

The problem of the relative abundance of electrolytes in the normal and abnormal tissue is apparently of considerable importance, but is surrounded with serious technical difficulties. The ratio of electrical conductance to depression of the freezing point, */A, is significantly higher in the sap from the wall than in that from the abnormal tissue. This indicates that the concentration of electrolytes is relatively higher in the fruit wall, while that of non-electrolytes is higher in the sap of the teratological structure.

Other evidences support the immediately foregoing conclusion. Thus, the fact that the percentage difference between the sap from the wall and that from the abnormal whorl of carpels is far larger in the case of conductivity than for osmotic pressure, or specific gravity, indicates that the differentiation of these two sets of tissues is greater with respect to electrolytes than to non-electrolytes. Again, the apparently higher mean molecular weight of the solutes from the abnormal tissues evidences in favor of the presence of greater quantities of non-dissociated substances.

In conclusion we should like to state that this paper is purposely limited strictly to the presentation of facts. That the statements concerning matters of fact are sound is, we believe, established by the high degree of consistency in the results of numerous physicochemical determinations based on very large masses of biological material.

We feel that in the present state of our knowledge an attempt to discuss the physiological significance of these results is premature. For the present, the imperative need of physiology is sound quantitative measurements of essential variables, and harm rather than good is done by premature theoretical discussions.

LITERATURE CITED

DIXON, H. H. and W. R. G. ATKINS (1913). Methods of extracting sap from plant organs. Sci. Proc. Roy. Dublin. Soc., n. S. 13, 422-433. Also, Notes Bot. Soc. Trinity Coll., Dublin, 2, 154-165.

GORTNER, Ross AIKEN and J. ARTHUR HARRIS (1913). On a possible relationship between the structural peculiarities of normal and teratological fruits of Passiflora gracilis and some physico-chemical properties of their expressed juices. Bull. Torrey Bot. Club., 40, 27-34.

(1914). Notes on the technique of the determination of the depression of the freezing point of vegetable saps. Plant World, 17, 49-53.

HARRIS, J. ARTHUR (1906). Prolification of the fruit in Capsicum and Passiflora. Ann. Rep. Mo. Bot. Gard., 17, 133-145. HARRIS, J. Arthur and Ross AIKEN GORTNER (1914). On the influence of the order of development of the fruits of Passiflora gracilis upon the frequency of teratological variations. Plant World, 17, 199-203.

(1914). Notes on the calculation of the osmotic pressure of expressed vegetable saps from the depression of the freezing point, with a table for the values of P for A=0.001° to Δ=2.999°. Amer. Jour. Bot., 1, 75-78.