Photographic Researches and Portraiture   291

Jacobs gave him help. The first paper. is entitled : "An Inquiry into the Physiognomy of Phthisis by the Method of Composite Portraiture'." It contains illustrations of 47 composites and of 113 individual portraits. There is thus a great wealth of material to judge by. Unfortunately, and probably unavoidably, the portraits are all small, the individual smaller than the composite portraits, and this, I venture tb think, lessens the accuracy of judgments based on comparisons of this illustrative material.

The question raised by Mahomed and Galton was whether there was any justification for a belief in a phthisical diathesis ; it is of course clear that such may exist without involving a phthisical physiognomy. It is further possible that if such a physiognomy exists, it might be produced by the action of the disease itself. The material consisted of 261 male and 181 female photographs of phthisical subjects between 15 and 45 years of age taken partly at Guy's Hospital and partly at the Brompton and the Victoria Park Hospitals for Diseases of the Chest. A schedule for each subject dealt with : Age-extent of disease (advanced, moderate, slight)duration of disease (chronic, over 3 years ; medium, 1 to 3 years ; brief, under 1 year)-hereditary taint (strong, some, none)-onset of disease (insidious, or preceded by severe haemoptysis, bronchitis, pneumonia, pleurisy, syphilis, gout, alcoholism). These classifications enabled composites' of various groups to be made. A s control two series of female patients, each fifty in number, and a series of male patients 100 in number, all suffering from diseases other than phthisis, were taken without selection.

When all individual phthisical patients were compounded without selection, a composite was obtained strikingly like the composite portrait of the non-phthisical : see our Plate XXXIV. If there be made selections of the narrow, ovoid or 'tubercular' faces from the phthisical and non-phthisical patients, or again of broad faces with coarser features from the two, groups, we again reach composites closely resembling each other. In other words both phthisical and non-phthisical patients contained representatives of the same two types. Further, of the non-phthisical women 15'/, gave the narrow ovoid face while only 11-6'/. of patients with phthisis presented it. Among males the proportion of narrow ovoids was only 13'/. in the non-phthisical patients and 19,0'/. among the phthisical on the best estimation. Taken altogether the phthisical cases showed 14-3'/. and the other than phthisical 14-0'/ ° of the narrow ovoid or so-called `tubercular' physiognomy. The `tubercular' physiognomy is therefore not more common among the phthisical than the non-phthisical diseased population. Mahomed and Galton write

"Let us here emphasise the fact that we are now comparing phthisis with other diseases, and not with the healthy population, and these observations would seem to show that a delicate person may fail in many ways besides being phthisical, and that a delicate narrow ovoid face

may mean liability to other diseases not necessarily tubercular." (p. 13.)

1 Guy's Hospital Reports, Vol. xxv, February, 1882.

2 In this paper a compound of composites is termed a co-composite, and if several co-composites are compounded the result is termed a co-co-composite. Composites and co-co-composites are positives, and require to be reversed before printing from them.

37-2