The agglutination is a specific phenomenon, but this specificity is only relative inasmuch as a typhoid serum may likewise exhibit a certain degree of agglutinating power on some strains of the colon organism. Similarly I have found that the precipitins are not absolutely specific. A typhoid serum which agglutinates a certain strain of the colon bacillus will at the same time produce a slight precipitate in a filtered culture of the same colon organism. Finally, I have observed that if an

more »

... be immunised with different bacteria at the same time a poly-agglutinating serum results, in the same way that a serum results which will produce a precipitate in filtered cultures of the several organisms used for immunisation. This had already been observed by Wolff as regards a poly-agglutinating serum. I next proceeded to study the precipitins which might be produced by the serum of animals that had been treated with liltered broth cultures of bacteria. Kraus, experimenting in this direction with the diphtheria bacillus, came to entirely negative results, and was led to the conclusion that there is no formation of precipitins in the serum of animals treated with germ-free filtered cultures. I may say at once that experimenting like Kraus with the diphtheria bacillus I came to a similar conclusion. But the results are of a different character if the experiments are extended to other bacteria. For example, cultures of the typhoid bacillus, colon bacillus, and staphylococcus aureus were freed from the organisms by filtration through a Berkefeld filter and the filtrates were injected into rabbits. In the blood of all the treated animals it was easy, after a certain time, to detect the formation of specific precipitins, as will be seen from the following experiments. EXPERIMENT I.-The filtrate from an old broth culture of the typhoid bacillus was added to three series of test tubes. To the first series were added varying quantities of the serum of a rabbit (from one cubic centimetre to one-twentieth of a cubic centimetre) which had been treated with filtered cultures of the typhoid bacillus. To the second series were added the same quantities of serum taken from an animal treated with filtrates of cultures of the colon bacillus, and to the third the serum of a rabbit which had received filtered cultures of the staphylococcus aureus. After being kept for 20 hours at blood heat two tubes of the first series contained distinct precipitates-viz., those to which one cubic centimetre and half a cubic centimetre of the typhoid serum had been added. The test tubes in the second and third series, on the other hand, remained perfectly clear. EXPERIMENT II.—An old broth culture of staphylococcus aureus was passed through a Berkefeld filter, and the filtrate added from a rabbit treated with staphylococcus filtrate, to a second half a cubic centimetre of serum from a rabbit treated with typhoid filtrate, and to a third half a cubic centimetre of serum from an animal treated with colon filtrate. At the end of 12 hours a precipitate occurred in the first tube and not in the other. EXPERIMENT III.-To the filtrate of diphtheria cultures I added in varying quantities the serum of a horse immunised in the usual manner with diphtheria toxin. I was, however, not able to detect any precipitin, although the tubes were kept at blood heat for 48 hours. The above observations show that on immunising animals with bacterial free filtrates of cultures of the typhoid bacillus, colon bacillus, and staphylococcus aureus it is possible to induce a formation of precipitins in their blood. The results were negative in the case of filtered cultures of the diphtheria bacillus. This fact, is in my opinion, an additional argument in favour of the intimate relation existing between the agglutinins and precipitins. For if we treat an animal with filtered diphtheria cultures we do not induce in its blood a formation of agglutinins (or only very slightly), but if we treat an animal, e.g., with filtered typhoid cultures, the blood will acquire a marked agglutinating power, though not so strong as if we had carried out the immunisation of the animal with unfiltered cultures. A final series of experiments was made in order to determine whether precipitins were formed in the serum of animals that had been treated with dialysed cultures of bacteria. For this purpose typhoid cultures were taken (45 days old) and dialysed by means of a Martin gelatin filter. The dialysates were injected several times into two rabbits, five cubic centimetres at a time being inoculated subcutaneously. 12 days after the first injection the agglutinating power of the blood of the animals was tested and it was found that it had acquired distinct agglutinating properties, a distinct agglutination of typhoid bacilli being produced by a dilution of 1 in 500. This confirms the observation made by Macfadyen that it is possible to induce a formation of agglutinins by treating animals with dialysed cultures. I tested the blood at the same time for the formation of precipitates, using as test material old filtered cultures of the typhoid organism. The results were positive. The serum of the animals treated with the dialysed cultures produced in doses of 0'5 and 0'25 cubic centimetre a precipitate in the filtered typhoid broth. The serum did not produce any precipitate in filtered cultures of staphylococcus aureus or the colon bacillus. My experiments, therefore, lead me to the conclusion that it is possible to induce the formation of precipitins as well as agglutinins by injecting dialysed cultures. My experiments lay no claim to completeness and it is my intention to continue and to expand them. Thus it would be interesting to test whether the serum of typhoid fever patients causes a precipitate in filtered typhoid cultures. The experiments which I have made point to the following conclusions : 1. The blood serum of animals treated with different preparations of natural albumins contains specific precipitins for the albumins. 2. The blood serum of animals treated with unfiltered bacterial cultures produces a precipitate in filtered cultures of the organisms in question. 3. The serum of animals treated with filtered cultures of bacteria likewise develops specific precipitins which produce precipitates in the filtered cultures of the same bacteria. 4. An exception was in the case of diphtheria cultures, the injection of which did not lead in my hands to the production of precipitins. 5. Animals which are treated with dialysed typhoid cultures develop the specific precipitins in their serum. 6. There is a close connexion between the agglutinins and the precipitins. I am greatly indebted to Dr. Allan Macfadyen for the help afforded me in the course of the experiments, as well as for the performance of the necessary inoculations. SENIOR SURGEON TO THE CITY ORTHOPÆDIC HOSPITAL, HATTON-GARDEN, E.C. THE main facts known about congenital torticollis have been described by many writers and therefore I propose to limit my remarks to a few practical points in the conduct of these cases. These points may be summarised as follows. 1. Division of the contracted sterno-cleido-mastoid muscle is usually required to cure the deformity. The anterior fibres of the trapezius and also deep bands of cervical fascia are also sometimes involved. 2. A retention apparatus is not usually necessary in the treatment of uncomplicated congenital torticollis, either before operation or as a means of retaining the head in a straight position after the division of a contracted muscle. 3. In operating we have to decide between an open wound and a subcutaneous section. The open wound (assuming that strict antiseptic precautions are taken) is desirable in the majority