In human schistosome infections there is always the possibility that a patient may harbour a preponderance of male parasites, or that female parasites may be absent or remain undeveloped for long periods of time. This possibility is not generally recognised, in spite of the fact that the above-mentioned conditions are frequent in experimental tests. And it is evident that, if such conditions do hold for man, the mere absence of eggs from the urine or faeces is of little avail in determining the presence of an infection or in estimating the effects of treatment. Since, further, the ova are not readily detected in faeces and slight infections of the urine are demonstrable only when the very last drops of urine are collected and centrifugalised, a less uncertain and laborious method of estimation is clearly desirable. The complement-fixation test is not entirely trustworthy, though a negative result and a normal blood count would prove useful evidence of successful treatment. It is suggested in this paper that the degree of eosinophilia may afford a reliable confirmatory test.