tion, and those which are bent towards the rent, are widest in the middle, and meet at the highest and lowest points of the rent, so the distances from the rent to which the strata are bent are greatest opposite its centre, and decrease gradually to its highest and lowest extremities. Again, as the sphere of the horizontal contraction has extended as far as the strata are bent, and as the distance from the rent to where the contraction in this direction commenced, is greatest in the middle of the rent, and decreases upwards and downwards from this place; so the distance which the strata have contracted horizontally is greatest at the middle, and decreases gradually towards the highest and lowest extremities of the rent.

From what I have seen of the smaller rents of this shape, in general when the distance f w is four yards, the distance dc is two feet, the distance v 1.90 yards, and the distance a b three yards. From these data the contraction horizontally is one-sixth of the distance in which that contraction took place. How much the contraction is perpendicularly we have no direct means of ascertaining; but that it is as much in this as in the horizontal direction is extremely probable; and on this supposition the distance which the strata have contracted less on the under than on the upper

side of this rent is one-fifth of their whole contraction in this direction.

The matter near this rent has contracted more horizontally in one part than in another. Thus its horizontal contraction is greater opposite the part 1, fig. 3, Plate XXIX., than opposite the part c; and more opposite the part d than the part e; and so on. The inequality in the matter's contraction in this direction is from onefifth to one-sixth of its whole contraction. This inequality gives the curvilinear bends to both sides of a rent, which necessarily produce inequalities in its width. But that arrangement of these inequalities represented by fig. 3 was formed by the inequalities of the matter's contraction taken perpendicularly as well as horizontally; because similar bends in the sides of this rent are never wholly opposite each other, but always situated lower on the upper side than on the under side; and this arrangement is owing to the circumstance before mentioned, that the parts of the strata on the former side, or the side g m, have contracted more perpendicularly or sunk lower than on the latter side, the side a f, of the rent. When the hollow parts on the upper side of a rent have sunk directly opposite similar hollows on the rent's under side, the variation in the width before described and represented by fig. 4, Plate XXIX., is produced; but corresponding hollows are not opposite each other; for the hollow a was originally opposite the hollow al, &c. There are many other variations in the ratio of the strata's bending, and in the regularity of the widths and positions of bended-tabular rents, which depend upon certain modifications of the principles already laid down. Some of these variations I will afterwards describe in separate essays.

Experiments to determine the Definite Proportions in which the

Elements of Organic Nature are combined. By Jacob Berzelius,
M.D. F. Ř.S. Professor of Chemistry at Stockholm.

(Continued from p. 409, Vol. IV.)

III. Analysis of the Ternary Oxides. 1. Analysis of citric acid. I shall begin with citric acid, because its composition is the most simple of all the ternary oxides, Ten parts of citrate of lead, strongly dried and decomposed by sulphuric acid, produced 8.9346 parts of sulphate of lead, equivalent to 6-582 of oxide of lead. Hence it follows that citrate of lead is composed of

34:18

100 Oxide of lead

65.82

190

Citric acid ...

100.00 These 190 parts of oxide of lead contain 13.588 parts of oxygen. In analyses made by combustion I found only 187 or 188 parts of oxide of lead combined with 100 acid. Citrate of lead retains humidity very strongly, and hence analysis makes the base always too small. As this analysis was one of the first that I made, I had not at that time found out the method of drying the substances in a sand-bath heated and exposed in a vacuum.

I made a great many experiments on the citric acid, because this acid and the tartaric served as a kind of exercise to prepare me for these experiments. A part of them was unsuccessful, because I had not at that time acquired the necessary practice in analysis. One among the successful experiments gave as the result of the analysis of 1 part of citrate of lead (which by combustion left for residue 64.956 per cent. of oxide of lead, and in which there was of course 35.044 per cent. of acid,) 0:1145 water, and 0.503 of carbonic acid in the state of gas; to this if we add 0.03 for the carbonic acid retained by the soda, we obtain 0-533 of carbonic ucid. This shows us that citric acid contains per cent. : Hydrogen

3.800 Carbon

41.369 Oxygen

54.831

100.000 We have seen that the capacity of saturation of this acid is 13585; but 13.585 X 4 = 54:34. This shows us that citric acid ought to contain four times as much oxygen as the base by which it is neutralized. When we turn these numbers into

volumes, we find that they agree perfectly with the idea that citric
acid is composed of equal volumes of its elements; that is to say,
it ought to contain per cent. :
Hydrogen

3.634
Carbon

41.270 Oxygen

55.096

100.000 If this be so, citric acid has the simplest composition of any of the ternary oxides. Its capacity of saturation indicates that it ought to contain four volumes of oxygen, four volumes of carbon, and four of hydrogen. Must we consider it as a compound of 12 atoms or of three? Experiments on the combination of citric acid with water seem to prove that it contains only three atoms, an atom of each constituent.

When we compare the result of my former experiments on the quantity of water contained in citric acid, we find that they do not coincide with the result of the analysis of citrate of lead. 10 parts of citric acid in crystals dissolved in water, neutralized by ammonia, and mixed with nitrate of lead, as long as any precipitate fell, produced 23 756 of dry citrate of lead. In this citrate there is 8.3 of citric acid; that is to say, that 100 parts of acid contain 17 of water: or that 100 parts of acid combine with 20.5 of water, which contain 18•l oxygen. But 18:1 X 3 = 54:3; that is to say, that in crystallized citric acid the acid contains three times as much oxygen as the water.

I took five parts of citric acid reduced to a coarse powder, and exposed them in a glass capsule exactly weighed to a temperature between 1180 and 122°. In 24 hours it lost 0.43 of its weight, and had assumed the form of a white light powder. In another experiment the loss was 0.424. No further loss was sustained by longer exposure to heat. This experiment shows that citric acid by efflorescing loses from 8:58 to 80 per cent. of its weight of water; which is obviously the half of the water of crystallization contained in the acid. Citric acid does not part with the rest of its water; for if we heat it to a certain temperature, it loses indeed weight, but at the same time it becomes brown, is decomposed, and leaves a brownish deliquescent mass, no longer possessing the properties of citric acid. In order to expel the water we must substitute another oxide in its place. From these experiments it follows, that in effloresced citric acid, or citrate of water, as it may be called, the acid contains six times as much oxygen as the water, and that in crystallized citrate of water the acid contains three times as much as the water.

This property of citric acid to allow itself to be saturated by water in another proportion than by the stronger oxides is peculiar; but it is easily explained by the simplicity of the composition

of citric acid, and cannot take place in other circumstances. The composition of citric acid may then be expressed by this formula, H + C + 0.

My expectation of being able to illustrate the composition of this acid by the examination of a subcitrate of lead was disappointed, by the property which the neutral citrate has of dissolving in ammonia, and of forming a triple salt, froin which the ammonia could not be driven off even in a vacuum.

Thenard and Gay-Lyssac, in their experiments on citric acid, obtained as a result 6:33 hydrogen, 33.811 carbon, 59.859 oxygen. This result differs very much from mine. Their analysis of the citrate of lime employed by them shows that very little water of combination remained in the salt. Besides, a correction for the water of combination remaining in the salt would only increase the difference between our results. I am unable to divine the cause of this difference.

2. Tartaric acid.--100 parts of tartrate of lead analysed by combustion, left 6:2.5 of oxide of lead in one experiment, and 62.48 in another. This differs but little from my old experiments, in which I found 62:2 per cent. of oxide of lead in this salt. It is composed then of Tartaric acid.

37.5

100 Oxide of lead

62.5

167

water

is ! Des

my

100.0 But 167 of oxide of lead contain 11.94 of oxygen. In one of

old experiments I found that 100 of tartaric acid are neutralized by 70-4 of potash, in which there are 11.93 oxygen. I have shown likewise in these experiments, that tartaric acid must contain five times as much oxygen as the base by which it is saturated.

l of tartrate of lead, equivalent to 0.5 of tartaric acid, produced in different experiments from 0·161 to 0.162 of water, equivalent to from 3.79 to 3.807 per cent of the weight of the acid of hydrogen. In the same experiments I obtained from 0.654 to 0.661 of carbonic acid, equivalent to from 35:36 to 35.98 per cent of the weight of the acid of carbon. The acid is then composed of Hydrogen

3.807 Carbon

35.980 Oxygen

60.213

100.000 59.7. If we examine the ratio of these numbers

, we find that the hydrogen is to the oxygen :: 66 : 1000; that is to say, that there is a volume of each. The carbon constitutes f of a volume. But if the oxygen is in reality five volumes, it is combined with five volumes of hydrogen and four volumes of carbon. Supposing then tartaric acid a compound of 5 H + 4C + 50, calculation gives its composition as follows:

cight

stitut

But 11.94 X 5

tható

Hydrogen

3.951 Carbon

36.167 Oxygen

59.882

ste

The

100.000 The true capacity of saturation of this acid is then 11.976, instead of 11:94 given by experiment. It follows from this, that in the neutral tartrates the carbon of the acid is combined with ly volume of oxygen, including the oxygen of the base. I have not been able to obtain a subtartrate of lead.

Let us now examine the result of the analysis of this acid made by Thenard and Gay-Lussac. They found it composed of hydrogen 6.629, carbon 24:05, oxygen 69.321; but they employed tartrate of lime, in which they supposed 77.577 of tartaric acid. That is to say, that they paid no attention to the water of combination of the citrate of lime. To find the quantity of this water I made the following experiments.

I dissolved neutral tartrate of potash in water, and precipitated it by muriate of lime; the neutrality was not altered. Hence it follows, that 100 tartaric acid saturate a quantity of lime containing 11.976 of oxygen ; that is to say, equal to that in the potash separated from the tartaric acid. I burnt 100 parts of this tartrate of lime in a platinum crucible, till the lime was reduced to a caustic state. To be quite sure of getting rid of all the carbonic acid, I allowed the lime to unite with a little water, and exposed it again to a strong heat; its weight was not altered. It weighed 21.64

. parts, and dissolved in diluted muriatic acid without the smallest disengagement of gas. 21:64 parts of lime require for saturation 50-55 of tartaric acid; hence it follows, that tartrate of lime is

; composed of Tartaric acid

50:55 Lime ..

21.64 Water of combination

27.81

a

100.00 But 21.64 of lime contain 6.094 of

oxygen,

and 27.81 of water, 24:54. Now 6:094 x 4 = 24:376 ; that is to say, that the salt contains a quantity of water of combination, the oxygen in which is four times that in the lime. It follows from this,

that what Thenard and Gay-Lussac considered as 100 tartaric acid, was in fact a mixture of 65.46 of acid with 34:54 of water.

If we subtract this proportion of water from their result, we obtain tartaric acid composed of Hydrogen

3.912 Carbon

36.889 Oxygen

..59.200

100.000 which agrees perfectly with the result of my experiments.