1915] on Anti-Typhoid Inoculation 435 



pushed to an amount which caused considerable constitutional dis- 

 turbance, while massive doses of vaccine may produce an initial 

 " negative phase." The effect was then tried of giving a second 

 moderate injection ten days after the first, so that the second stimulus 

 might be summated upon the first. The efifect is shown by the wave 

 sweeping up to 300 times the normal. A succession of doses at 

 ten-day intervals would give still further protection, but a balance 

 must be struck between the attainable protection and the incon- 

 venience involved both to the individual and to the unit in the midst 

 of miUtary training. It is considered that the double dose strikes a 

 happy mean. 



The need for, and interval between, the two injections has been 

 dealt with fully, because of the many questions which are constantly 

 being asked about it, and because of the great utility of this chart 

 w^hen hung up in the re-inoculating room. There are very few men 

 indeed who refuse their second dose when the chart has been shown 

 them and fully explained. 



After the third week there is a slight but rapid fall for a few 

 days, and then the agglutinative power slowly recedes to the normal 

 during a number of years. It is probable that normal is not reached 

 for five or six years, but it is considered that protection should not 

 be regarded as effective after the end of the second year, at which 

 time it is advisable that men should be re-inoculated. 



This has an important bearing on statistical data, as will be seen 

 when we come to consider whether a man who has been inoculated 

 more than two years previously should be included in the protected 

 or the unprotected groups. 



(b) The remote Results. 



The remote effects of inoculation upon enteric incidence can only 

 be estimated by statistical methods, with regard to which one factor 

 calls for preliminary comment. 



It is that — just referred to— of the classification of partially 

 protected men, i.e. those who have had only one injection or who 

 have been fully inoculated more than two years since. To include 

 them among the fully protected would raise the enteric incidence 

 in this group, while to place them among the wholly unprotected 

 would reduce the enteric incidence in that group — either of which 

 does injustice to the results of inoculation in the comparison of 

 relative incidences. The only sound course is to place these men in 

 a group by themselves. 



With this initial remark one turns to the available statistics, an 

 these have now become so numerous that only a few of the mor 

 striking or interesting can be discussed. Three, of an increasing 

 relative importance, are selected as indicating the results which have 

 been obtained in our own army. 



