INTRODUCTION. 



The following study on tubal pregnancy is to be regarded somewhat as ;i by- 

 product from our embryological collection. Originally we were of the opinion that 

 the very earliest stages of human development would be found in a recent tubal 

 pregnancy removed by the surgeon. Before 1897 physicians were under the impres- 

 sion that an embryological collection should consist only of normal specimens, and 

 the first tubal pregnancy added to the collection (No. 109) was sent to us by Dr. 

 Gushing because it contained a normal embryo. We soon found that we would not 

 obtain very early specimens by this method from the fact that the diagnosis of 

 tubal pregnancy is made much too late. The smallest normal specimen we have 

 received (No. 808) contained an embryo 6 weeks old. The very small tubes which 

 have been sent invariably contained pathological embryos or small ova without 

 embryos. According to our records 59 per cent of tubal pregnancies fall in the 

 latter class. In 46 carefully selected cases from Dr. Kelly's clinic this percentage 

 is raised to 84.8; if we could collect all cases of tubal pregnancy, the probabilities 

 are that it would be still higher. 



The normal specimens accumulated slowly among the numbers of our collec- 

 tion below 500. In this group 21 out of 27 specimens, which had been examined by 

 physicians before they were sent to the laboratory, contained normal embryos. 

 Among the specimens that had not been previously examined, only 4 out of 19 were 

 normal. (See table 1.) 



TABLE 1. List of specimens reviewed in this publication (total number 117). 



I have been able to collect three sets of statistics regarding the frequency of 

 normal and pathological embryos and pathological ova in tubal pregnancy. The 

 first includes specimens examined by physicians who sent them to us; to the second 



5 



