she had a very severe attack lasting four and a half days, during which I was able to see her twice. The paroxysm had seized her suddenly in the night while at rest, but she was able to continue at work the following day ; after that, however, she had to take to her bed. When I saw her she was lying on her left side, very much exhausted and sweating considerably. She complained of pain in the upper part of the chest and down the left arm, but it was not acute. The pulse was 180, but in other

more »

... pects the heart was normal. When seen two days later she looked extremely ill and there was an ashen pallor of the face ; she did not complain of pain but was feeling very greatly exhausted. The pulse was uncountable at the wrist, being 260 with the stethoscope, and there were signs of dilatation of the heart. The next day the pulse dropped to 90 and she was soon out of the attack, but she felt weak and battered for from three to four days, and then resumed her work. This was by far the longest attack she had ever had and it has not been repeated. During the attack digitalis proved of no avail ; she seemed to experience relief by having cold water poured over the wrists. Since she has been under out-patient treatment the attacks have certainly been less frequent. The drugs which have seemed most useful are bromide of potassium and hyoscyamus. CASE 2.-The patient was a young man, aged 23 years. When 18 years old he was said to have had "muscular rheumatism" and was in bed for ten weeks, so presumably the valves of his heart ware damaged at that time. A year later he had an attack of influenza, followed by pleurisy and pneumonia. About two and a half years ago he began to have attacks of palpitation of the heart, which at first occurred about once a month, but for the last one and a half years he had had on an average two a week and had never been for more than a fortnight without an attack. The first attack seemed to come on after rather a long bicycle ride, but he had never been able to discover any other cause for the ensuing attacks. In appearance he was a small, rather wirylooking fellow, but pale-faced with an anxious expression and dilated pupils. On examination his heart showed some hypertrophy of the right ventricle ; there were a presystolic thrill and murmur, and the liver was two inches below the costal margin. As a rule the pulse was rather slow, varying from 60 to 66, full, and of moderate tension. The blood pressure was 120 (taken by Martin's manometer). The attack of palpitation came on suddenly and the pulserate ran up to 240, becoming uncountable at the wrists but the beats were regular and the ordinary condition of his heart was unchanged ; he sometimes had a feeling of impending death but there was no acute pain ; he usually sweated profusely. This condition lasted from 20 to 30 hours and then came to an end as suddenly as it began. Like Case 1, he said it felt as though the heart had jerked itself into the right place again. He generally felt very drowsy after an attack. Often the attacks began at night, when he was lying quietly in bed, and these were usually the worst. At times he felt particularly well for a couple of days prior to the attack. Though the attacks certainly caused him more general distress than Case 1, he was yet able to get about while they lasted and had indeed come up to the hospital with an attack on him. He came into the hospital for two months, but the rest in bed did not seem to do him much good and treatment during the attacks was not satisfactory, although they were less prolonged than when at home. Among drugs erythrol tetranitrate seemed to give him most relief and occasionally digitalis cut short an attack ; neither drinking hot coffee nor ice to the praecordium nor a blister to the spine appeared to be of the smallest use. On several occasions, however, he found that vomiting relieved him, sc acting on this he was given an emetic and nearly always the attack ceased after he had vomited. This method of treat. ment certainly had a depressing effect upon him in the intervals and is not to be recommended. From pressure on the vagi in the neck, which had been suggested by some, I obtained no satisfactory result. Rosenfeld claims to have cured four cases by making the patient press bis elbows into the sides, and at the same time making pressure on the abdomen. This certainly afforded some temporary relief in this case, the patient saying that during the attack the principal discomfort was the feeling of shaking in the abdomen, which extended down to the thighs, and this was relieved by the pressure. There is a,good deal of literature on this subject, but it can hardly be described as illuminating, either as regards cause or pathology. There is some agreement in regarding over-work, either bodily or mental, or an injury as the exciting cause, but this did not appear in the above cases. As to pathology, it would seem natural to suggest a lesion of the vagus, but Professor Clifford Allbutt has pointed out that this would only raise the pulse to 140 and certainly would not account for the great acceleration of the rate described above. Others have regarded it as a bulbar spinal neurosis, while S. West 1 considers the myocardium to be the seat of the lesion. Gibson has pointed out the analogy between paroxysmal tachycardia and delirinm cordis, which would accord with the idea that the seat of the lesion is in the myocardium. Such post-mortem examinations as have occurred show no trace of any nerve lesion, whereas fatty degeneration of the heart muscle has been found once, chronic interstitial myocarditis twice, and cardiac dilatation three times. Hoffman 3 states that in 13 necropsies myocarditic changes were present in all, and these were usually recent. He does not, however, believe that the myocardial condition was responsible for the disease. On the contrary, he is of opinion that the essential lesion must be located in the central nervous system and considers the paroxysms similar in nature to an epileptiform attack ; certainly there is much in these attacks to suggest something in the nature of Liveing's "nerve storms " or of the paroxysmal neuroses. Some cases present a great variety in the paroxysms which at times last only a few hours, at others for several weeks. Hochhaus 4 quotes a case in which there was cedema of the whole body, bronchitis, and anuria, which all vanished rapidly when the heart returned to its normal state. Post mortem nothing definite was found. According to Herringham if the attack lasts more than five days secondary symptoms-e.g., bronchitis and pulmonary oedema -will probably appear. Bouveret,6 who was the first to give any complete account of paroxysmal tachycardia, does not include cases with organic cardiac lesions, but this is surely an unnecessary refinement, for paroxysmal tachycardia has clearly no essential connexion with valvular disease and, therefore, although Case 2 was suffering from mitral stenosis, it seems quite reasonable to regard it as an example of paroxysmal tachycardia. The prognosis is certainly not good, for after 30 years of age a patient is never safe from death. In the intervals, however, patients are able to resume their ordinary avocations and indeed to do fairly hard work without experiencing either palpitation or dyspnoea. According to Bouveret, the acute dilatation of the cardiac cavities which characterises a prolonged attack (acces de longue duree) does not pass into a chronic dilatation which can be appreciated in the intervals of the paroxysms. Nevertheless, the attacks tend to recur with greater frequency, gradually becoming longer, more intense, and more grave till the final syncope. now become a valuable aid to diagnosis in many forms of disease. When one begins to study the pulse and the circulation in fevers, the changes which are attributable to the fever itself must be carefully distinguished from those which are merely secondary to pyrexia. The increase in the pulse-rate which occurs is-in the first case, at any ratesimply due to the increase in temperature of the blood. Liebermaster has shown that in man there is an increase of about 14 beats per minute for each degree Centigrade-or eight beats for each degree Fahrenheit-of increased temperature. This phenomenon is particularly well illustrated in the early stages of many cases of enteric fever. Increase in the pulse-rate beyond this physiological ratio must be 1