before turning to them specially we will give a brief summary of the results in the three batches considered collectively. In all thirty-five mice were inoculated bilaterally and in twenty-two of them bilateral tumours developed. Thirteen mice were submitted to operation. Spontaneous disappearance did not take place in any one of the nine tumours not submitted to operation. Of the thirteen mice operated on, ten lived until they were killed at the completion of the experiment. In one only of these ten did the tumour disappear completely. This is depicted in fig. 5 as (2). In one other mouse, the tumour operated on disappeared, but the control tumour on the opposite side also disappeared. This is no. 4 in fig. 5. In the other eight cases there was ultimately renewed growth; but in all, the tumours on the side submitted to operation remained much smaller than the control tumours, although, as will be seen by a reference to the chart (fig. 5) the larger of the two tumours was operated on. The reasons for continuation of growth in seven cases were muscular infiltration in four, and the compensatory enlargement of some minute vessel not recognised and occluded at time of operation in three. Fig. 5 is a record of the results of operating on five mice of the batch 95 E. After what has been said a detailed description of this experiment is unnecessary. The reason for continued growth in mouse (1) was that the pectoral muscle was infiltrated and the anterior part of the tumour got a blood supply through muscular twigs. It will be noted in the case of mouse (2) that only three weeks elapsed from time of operation until the mouse was killed off owing to the large size of the control tumour on the opposite side. Hence it is impossible to regard the disappearance of the tumour as a cure. The disappearance of the tumour in mouse (4) must be looked upon as a case of spontaneous absorption. In mouse (5) the persistence of growth was due to enlargement of a small lumbar vessel not seen at the time of operation. II. It is unnecessary that we should deal at length with the results obtained. in the case of the adeno-carcinomata which have a tendency to hæmor. rhage and the formation of cysts containing blood. Had we found * In working with Jensen's tumour there is this objection to using a control tumour in the same animal, that owing to the comparatively rapid growth of such tumours the observation time is very limited. 6. 7. 32 23 F 32 23 F FIG. 6. Diagrammatic representations of growth of tumours. The batch from which the tumour came is printed before the graphic representation of its growth. Nos. 1, 2, and 3 were hæmorrhagic adenocarcinomata (tumours 39 and 50), nos. 4-7 squamous-celled carcinoma tumour 32. The mice with the tumours in situ are first represented as charted immediately before operation. Following operation the condition of the tumours was charted at intervals of 8 days. In all cases save no. 1 the tumours were on both sides and whilst one side was operated on the tumour on the other side was used as a control. In the case of no. I there were two tumours on the same side of the body but quite distinct. The blood supply to the anterior one was tied, the tumour disappeared, but later renewed growth occurred. The posterior tumour steadily enlarged. Nos. 2 and 3 are shown as a contrast: the left side was operated on in both cases, but whereas in no. 3 the tumour disappeared entirely, in no. 2 it took on a renewed growth after an apparent complete disappearance. In no. 4 the left side was operated on but the tumour on both sides disappeared. In no. 5 the left side again was operated on, the tumour disappeared entirely whilst the control tumour first increased in size, then decreased, and then again took on renewed growth until it reached a relatively enormous size. No. 6 had the right side operated on, the tumour on this side disappeared whilst the control tumour steadily enlarged. In no. 7 the right side was operated on but the operation produced little effect and at necropsy the operation was found to be incomplete. evidence that there was any pronounced difference between the reaction of these tumours and that of the preceding type of growth, we should have set out the results in full. As it is we are justified after operating in 15 cases in stating that the same results are found in the cases of the growths tending to form hæmorrhagic cysts as in the more solid and less vascular alveolar carcinoma. Our first series of experiments were made with a batch of tumours 39/M. These tumours belonged to the first grouping, i. e. unilateral tumours; three of these tumours disappeared entirely, and there were no signs of recurrence some weeks later; in the case of the fourth renewed growth followed and persisted. Complete disappearance of 3 out of 4 of these tumours at first suggested that they might be peculiarly susceptible to interference with their blood supply, but a more complete study in a larger number of cases, especially with bilateral tumours in the manner set out before, showed that no such susceptibility was really present. On the chart forming fig. 6 we have represented the results following operation in two cases of tumours from 50*4/F. In this batch, 3 mice grew bilateral tumours out of a total of 24 mice inoculated. These three mice were operated on, the blood supply to the larger tumour being exposed and occluded, the result being that in one the tumour completely disappeared and in two renewed growth occurred. On the same chart we have pictured the result of interference with the blood supply to the anterior of two tumours situated on the same side of the body in a mouse. The figures show the way the tumour thus dealt with apparently disappeared entirely but ultimately recurred again, whereas the posterior tumour steadily grew until it reached a relatively enormous size. At the postmortem examination in this case secondary growths were found in the lungs. III. In considering the results obtained after occluding the blood supply to tumours resulting from the propagation of a squamous-celled carcinoma it is necessary to mention a point which is of some importance. This is, that this tumour often infiltrates the neighbouring structures quite early, and if it has already invaded the muscular wall of the abdomen or thorax, successful operation is out of the question. In consequence of this infiltration operation has to be carried out when the growth is quite small, and when it can therefore be very easily separated and all its connections studied and divided without any trouble. It is Described in detail in this Report. possible, therefore, that in the majority of operations carried out with this tumour the occlusion of the blood supply was more complete than is usually the case. We feel that before attaching any importance to the rather obscure results following some of the operations in this variety of growth, we should like to make a further investigation. In operating, that tumour was exposed which was thought to be free, i. e. not fixed to muscle by infiltration. This often meant that the smaller tumour was the one to which the blood supply was occluded, thus differing from the practice in the case of the alveolar carcinoma and hæmorrhagic adeno-carcinomata. In all 7 operations were completed, only bilateral tumours being used. In two of these the tumours on both sides disappeared, and no recurrence has taken place over two months later. One of these is charted in fig. 6. In one case the tumour on the side operated on disappeared, whilst the tumour on the unoperated side grew larger, then ulcerated, decreased considerably in size and then again grew to a large size. This has also been figured (fig. 6, no. 5). In one case no recurrence had taken place 16 days after operation, but there had been very little increase in size in the control tumour. In one there was no recurrence on the side operated on, whereas the control tumour grew to a weight of 9.2 grams in a mouse of 17.2 grams. This case is charted as no. 6 in fig. 6. In two cases the tumour operated on rapidly recurred and grew to a large size. One of these is shown as no. 7 in fig. 6. Similar observations have been made with a transplantable spindlecelled sarcoma described in detail by Dr. Haaland on a later page. With sarcoma the same difficulties are encountered as have been met with for squamous-celled carcinoma, and the results of operating on the tumours in the way described have given no more hopeful results as yet. From this summary it will be seen that whereas the percentages of successful operations appears high, this is only apparent and unsupported by the closer investigation which reveals a tendency to spontaneous absorption on the part of the batch as shown by complete disappearance of both tumours in two cases. When this feature of the batch as a whole is considered together with the peculiar conditions, as mentioned above, under which operation is carried out, we feel that a more extended investigation of this variety of growth would justify the conclusion that in the absence of a tendency to spontaneous cure any imperfect operation will be followed by persistence of growth. In concluding this summary of experimental work, we trust that the evidence brought forward may be of some use in estimating the possible |