As illustrating the character of the temperature curves obtained from tuberculous patients we may take those obtained from patients treated at Addenbrooke's Hospital, at the Cambridge County Tuberculosis Dispensary, in sanatoria, or at one or other of the open-air colonies founded near Cambridge. We select these as here the patients could be kept under careful observation and any special treatment, reaction, and results noted at once; this allowed of their importance being properly appreciated. In Fig. 21 we have an exact reproduction, on a reduced scale, of one of the record charts with notes as made at the time. It is obvious that, on account of the space required, it would be difficult to reproduce these charts in extenso, and as the record is still exact even when subjected to lateral contraction, all the charts have been drawn to scale but with this lateral condensation.As the records of rectal temperatures are reliable and are now those usually noted in most sanatoria we decided that, for the present, it would be well to utilise records of this type only. The flgures inserted on the " hour " lines of the charts indicate the mouth temperatures taken with a mercurial clinical thermometer, and the times at which they were obtained, and are here given for purposes of comparison and control. The letters are merely indices of certain events of the day. D. = dinner, T. = tea, Mic. = patient rose to micturate, S. = supper, B. = breakfast, Sm. = smoking, &c. {See Fig. 5, p. 285.) *Parts I., II., III., and IV. were published in THE LANCET of Jan. 22nd (p. 173) Feb. 5th (p. 281), Feb. 12th (p. 338), and Feb. 26th (p. 450).On p. 285 are given temperature records (Figs. 7 and 8) obtained from a couple of patients suffering from tuberculous disease. The patients remaining in bed in Addenbrooke's Hospital had their temperatures recorded at the Pathological Laboratory a third of a mile away. These excellent records show how completely the external influences that might affect the minute electrical currents utilised have been overcome, and how sensitive is the apparatus.In the disease selected for our experiments the temperature is often continuously high, but fluctuating. Consequently it was necessary in designing and working the apparatus to allow for a sensitiveness much below that usually employed, though far greater than any that can be obtained by means of an accurately standardised and rapidly acting clinical mercurial thermometer. The first thing that strikes one in these records is that the curves, though profoundly altered, the temperature varying greatly from hour to hour, still maintain in a remarkable degree their fundamental diurnal variations.In the case, J. S., from which Fig. 7 was taken, the effect of a dose of tuberculin (0'001 mg. of Koch's bacillary emulsion), which produces a distinct rise of temperature almost immediately after its administration, is most striking and interesting. The patient at the time the record was taken was extremely ill, the clinical signs in the lung being those characteristic ...