Two series of observations have been made. One experiment consisted in feeding a large number of rats on an exclusive ox-flesh diet, this being essentially a mixed diet of protein and fats. In this experiment it was observed that the percentage weight of the liver in meat-fed rats was, as a rule, considerably greater than in the control bread-and-milk-fed subjects, the difference being such as could not readily be explained by mere variation in the blood-supply of the organ. An objection may be raised that the use of an exclusive flesh diet is too unnatural a regime to allow any deductions to be drawn which can be applied to conditions obtaining in man. To meet this objection attention is therefore more particularly directed to the detailed results of a second series of observations, which consisted in comparing the percentage weight of liver tissue in wild rats living, it is assumed, on a highly nitrogenous dietary with those of an equal series of animals obtained chiefly from the same source which were fed on a diet of bread and milk in fixed proportions - a diet which contained a much smaller proportion of nitrogenous food. This diet of bread and milk is one which has been proved to be well adapted for the growth, general health, and fertility of the ordinary tame rat.

Details Of The Experiment. Young Rats

A series of nineteen young wild rats were caught in the basement of a large hotel. It may, I think, safely be assumed that the diet of these animals was a mixed one, containing a considerable amount of nitrogenous foodstuffs. The coats of many of the animals were unusually rough and dry, but otherwise the animals appeared to be in good health. Nine of the rats, of weight ranging from 72 to 150 grammes, were at once killed, ordinary coal-gas being employed as the lethal agent; the remaining ten were fed in the laboratory for ton works on the ordinary laboratory diet bread and milk, and were then killed in the same manner. The livers were weighed out of a; per cent formalin solution, the excess of fluid having previously been removed. Table l. gives the record of the weights of the animals and the percentage weight of the liver in the two series.

1 Chalmers Watson, Lancet vol. ii., 1907.

Table I

SERIES 1. Young Wild Rats Killed on Receipt.

Weight of Animal in Grammes

Weight of

Liver.

72

5.6

72

6.9

75

5.5

95

6.5

96

6.7

106

6.6

130

60

142

5.3

150

5.2

Average ... .

6.0

Series 2. - Young Wild Rats Killed after Ten Weeks' Bread and Meat Diet.

Weight of Animal in Grammes.

Percentage

Weight of

Liver.

no

3.6

120

3.5

120

50

125

47

130

2.9

132

60

142

3.9

150

3.2

150

4.0

170

37

Average ... .

4.0

The figures show that the average percentage weight of the livers of the nine rats which were killed on receipt and which are assumed to have lived on a mixed diet containing much nitrogenous food was 60, whereas that of ten rats obtained from the same source and at the same time, which were subsequently fed for ten weeks on a bread-and-milk diet, was 4.0 grammes.

Fully Grown Animals

Thirty fully grown rats were obtained from a number of sources, the source in each instance being given in Table II. As in the case of the younger rats, it was noted that the coats of many of the animals were rough and dry, but otherwise the animals appeared in good health. Fourteen of these were at once killed, the remaining sixteen were fed in the laboratory for periods ranging from nine to eleven weeks. The weights of the animals and the percentage weight of the liver in the two series are given in Table II.

The figures in Table II. show that the average percentage weight of the livers of fourteen adult rats which were killed on receipt was 5 27 grammes, whereas that of sixteen rats obtained from similar sources and at the same time, which were subsequently fed on a diet of bread and milk for periods ranging from nine to eleven weeks, was 3.87 grammes, which represents a striking difference. There are two points in Table II., Series 4, which call for comment. The weights of the laboratory-fed rats show a striking uniformity in the series, the range of variation being a small one. On the other band, the figures in Table II., Scries 3, show that two animals form a striking exception to the general rule in this series, the average percentage weight of the liver in these two rats being actually less than in any of the bread-and-milk-fed subjets. It is noteworthy that these two animals came from a bud shop, which renders it probable that their diet consisted mainly of grain foods. (It was also observed that in these two animals the histological appearances of the thyroid gland were markedly different from the others in this series.) This result, then, confirms that obtained in the case of young rats given in Table I., and, assuming the correctness of the statement that the diet of the animals in their natural state contained a considerable amount of nitrogenous food, the facts indicate that the livers of adult rats which are fed on a diet containing much nitrogenous food have a considerably higher percentage weight than those of animals fed on a diet containing a smaller amount of nitrogen.

Table II

SERIES 3

Full-grown Wild Rats Killed on Receipt.

Weight of Animal in Grammes.

Percentage

Weight of Liver.

Source.

190

6.15

Sewer.

190

6.94

"

200

6.15

"

207

6.13

"

220

5.23

Canal bank.

237

5.57

Sewer.

240

4.97

"

242

...

...

260

5.83

Hotel basement.

260

4.05

"

270

3.25

Bird shop.

305

5.17

Hotel basement.

342

5.11

Canal bank.

370

2.78

Bird shop.

Average .

5.27

...

SERIES 4

Full-grown Wild Rats after Feeding on Bread and Milk from Nine to Eleven Weeks.

Weight of

Annual in

Grammes.

Percentage Weight of Liver.

Source.

162

3.93

170

3.76

180

3.83

All from

190

3.57

200

3.0

hotel

205

3.80

225

4.04

basement

237

4.38

260

4.30

or

270

3.55

275

3.78

canal bank.

290

3.75

300

4.06

320

4.31

Average .

3.87

...

In arriving at this conclusion I am aware of the importance of having regard to the possible operation of factors other than the highly nitrogenous diet in inducing the greater amount of liver tissue present in wild rats which were killed on receipt. Of these the two chief are, 1 believe, (1) muscular exercise, and (2) the admixture of foodstuffs. That the first is a factor of importance is rendered improbable by the fact of the low percentage weight of the liver in the two rats in Table II., Series 3, which were caught in the bird shop and killed on receipt. With regard to the second, I think it probable that the amount of fat and carbohydrates ted in the mixed diet is a factor of very considerable importance, but in the circumstances it is not possible, nor is it necessary, to dissociate these from the more important nitrogenous foodstuffs. While keeping in view, therefore, the possibility that other factors, both in the diet and in the animal's environment, may require to be considered, I believe that the difference in the amount of liver tissue present in the two series of animals is chiefly dependent on the relative proportions of nitrogenous foodstuffs in the dietary, and we have in these results a confirmation of those referred to in an earlier part of my paper as to the influence of an excessive meat diet on the liver.

It must not be supposed that I am assuming that the larger percentage weight of the liver tissue in the wild rats killed on receipt is to be regarded as pathological. No such assumption is made. The alternative view, that the smaller percentage weight of liver tissue in the bread-and-milk-fcd animals is pathological, may appear to some as the more probable one. But this assumption would be equally unwarranted. Indeed, since it was a noteworthy feature in the experiment that within a week or two of the onset of bread-and-milk feeding, the coats of the rats invariably lost the rough, dry aspect referred to in the text and became smooth and sleek, the balance of clinical evidence is in favour of the bread-and-milk regime. But it is not essential to elaborate a comparison. The one important fact to which attention is directed is the evidence that the liver is modified by diet, and, in particular, that a mixed diet which contains a large amount of nitrogenous food throws a greater strain on the liver than a diet, given in unrestricted amount, in which the nitrogenous elements are in smaller proportion.