SUMMARY. 103 



The variation in the percentages showing infiltration of the decidua is consid- 

 erable, ranging from 45.5 to 100 per cent. In the first three and in the sixth 

 group the percentage ranges from 90 to 96. It is 100 per cent in the fourth 

 group and 46 and 50 per cent in the fifth and seventh groups, respectively. There 

 is also considerable variation in the incidence of slight as compared with intense 

 infiltration. The former ranges from 11 to 86 per cent, and its incidence is approxi- 

 mately comparable in the first, second, and third groups, in which it ranges 

 from 11 to 19 per cent. Slight infiltration was most frequent in group 5, that of 

 cylindrical embryos, where it was present in 86 per cent. Its incidence was quite 

 comparable in the stunted fetuses (group 6), but very low in group 7. The rela- 

 tively high incidence of intense infiltration in this group, unless largely secondary, 

 is surprising, in view of the fact that many of these specimens have been retained 

 so long and are so degenerate. That intense infiltration is so frequent in group 

 1 is not surprising, and its presence likely leads to an early termination of the ges- 

 tation, no matter whether the endometritis existed previously to the gestation or 

 was incidental to or consequent upon interference. 



The occurrence of such a decided variation in the presence of infiltration of 

 the decidua would seem to be the best possible evidence that this infiltration has a 

 pathologic significance. Were the leucocytes, upon the presence of which the 

 opinion regarding infiltration was mainly based, a constant constituent of the 

 normal decidua, they should have been present in all specimens. This could fail 

 to be true only if this infiltration occurred in the earlier months of pregnancy 

 alone. 



A slight increase in the frequency of the occurrence of hydatiform degenera- 

 tion is seen in the first three groups, in which it rises from 69 to 85 per cent, being 

 lowest in group 5 and next lowest in the last group. The gradual increase 

 in its presence in the first three groups is easily explained by the fact that some 

 time is necessary for its development, and that it is difficult to recognize it in its 

 early stages or in macerated fragmented villi. The decline in its occurrence in the 

 last groups may be due to abortion of the conceptus soon after the development 

 of the degeneration. Its relatively extreme rarity in full-term births can be 

 explained similarly. The most interesting thing in this connection is the fact 

 that the decidua was infiltrated in every instance in which it was included in 

 the cases of hydatiform degeneration occurring in the first four groups. Since the 

 last three groups contained so few specimens, the percentages for each group are 

 of very little value, but the presence of infiltration seems to be far less common 

 in the specimens in these groups. This may be due to the fact that some low-grade 

 infections spontaneously disappear, but not before they have seriously damaged 

 the conceptus, which may then be retained for some time. 



The occurrence of changes suggestive of lues rises gradually from zero in 

 the first group to 10 per cent in groups 3 and 7, the incidence being roughly com- 

 parable in groups 4, 6, and 7, although group 5 contains only about half as many 

 as these three groups. 



Tubal. Approximately 73 per cent of all the tubal specimens classed as 

 pathologic in the first 1,200 accessions fall into the first two groups, which 



