A further point, which can be cleared up by the statistical method, is the question whether the fact that the glands never came to functional development, is to be considered as favorable for the development of tumors; this, it seems to me, is far too often asserted a priori, without statistical proof. Statistics disprove this assumption. Out of 236 female patients observed by me with mammary carcinoma, 23 were single and 213 married. The number of single patients is thus very small (9.74 per cent.). Velpeau found in 213 cases, 25 unmarried (15.34 per cent.). With such statistics based on hospital patients, we cannot take " un married" as identical with " childless; " but though we wish to know how much influence, not only the absence of sexual excitation, but also the want of full physiological development of the breasts, has upon the origin of cancer of the breast, the material for these statistics is far too uncertain; whether this has any influence on the unimportant dissimilarity of Vel peau's and my statistics, I cannot say. At all events, such numbers do not prove much more than that the unmarried are not free from carci noma. If we wish to be more exact, we must know how many un married and how many married lived under observation during a certain period of time, and thus fix the relative numbers; then we could say of the unmarried X years, X per cent. were affected, and of the married, X per cent., etc. As the numbers are now nearly the same, we cannot dispute the assertion that they sufficiently represent the proportion of the unmarried to the married during the period of life in question, and may thus prove as well that carcinoma is just as frequent in unmarried as in married women.
Similar objections present themselves to the explanation of the number if we compare the number of fruitful and unfruitful married women suffering from mammary cancer. Velpeau noted 138 cases, of which 28 (13.04 per cent.) were unfruitful and 110 were fruitful; I have observed 209 cases, of which 23 (11 per cent.) were unfruitful and 186 were fruit ful. How much depends upon the size of the numbers here, is shown by the statistics of Al. von Winiwarter, who has used some of my cases (170) in his statistics; he calculated the number of unfruitful women with mam mary carcinoma at 16.2 iper cent.; it should be mentioned here that of the thirty-nine women whom I observed further only one was unfruitful. The difference varies in the third statistical calculation, in regard to the unfruitful women between 11 and 16.2 per cent.; at all events it is not considerable. These numbers show indisputably that a greater number of fruitful than unfruitful women suffer from mammary carcinoma; but they do not show that fruitfulness is an important factor of the disease, since the number of fruitful women is enormously greater than the num ber of unfruitful ones. If these numbers accurately represent the relation
of the two categories to each other, then each is equally disposed to car cinoma.
I will now give the figures with regard to lactation. Velpeau noted: Of 110 fruitful women with mammary carcinoma, 60, or 54.54 per cent. had nursed; 50, or 45.46 per cent. had not. Al. von Winiwarter noted: Of 102 women with mammary carcinoma 37, or 36.2 per cent. had never nursed; 65, or 63.8 per cent. had nursed their children. This, probably, can only serve to show what is already known, that the French women more seldom nurse their children, or are more seldom in a condition to nurse them, than the Austrian women. That, therefore, more suckling women in Austria must be affected, is clear; and further, since the num ber of nursing women in female hospital patients is always larger than that of those not nursing, the excess of the former is not astonishing; the figures do not furnish indisputable proof that lactation predisposes to mam mary carcinoma. Certainly these relations are hard to express in figures; for if lactation is to enter into the question of the formation of carcino ma, it will be of interest to know whether it has been soon and suddenly interrupted, or unduly prolonged, whether many or few children were nursed, at what intervals, etc., also the number of abortions or premature labors, etc.; and thus the question, apparently so simple, is dependent upon so many individual observations, that it is not yet worth while at tempting to group them for statistical purposes.
That hardenings in one or both breasts during lactation, remaining after the weaning of the child, and after the retrogression of the glandular function, grow and develop into tumors, perhaps sarcoma or carcinoma, is considered beyond doubt by many authors; but this rests almost exclu sively upon the statements of the women, not upon the physician's obser vation, and ig therefore to be accepted with caution.
It is certainly no pleasant task to analyze so critically the results of careful and arduous labors, and much less so, when one has himself taken the warmest interest in such work. I am far from undervaluing the worth of such statistical researches for clinical exposition; great progress has been made therein, compared with the former foggy statements re garding such relations. But we cannot be too cautious when we use such researches as supports for these or those hypotheses which clear up the origin of tumors of the breasts, and eventually of all tumors. If an hypothesis is to have value, its supports must be irreproachable. That this is not the case with regard to the reiterated assertion that the mam ma is especially predisposed, by the peculiar nature of its development and by its periodic function, to the formation of tumors is probably ap parent from the above argument.