In 1964, while I was working as codirector of the University of Colorado's (Boulder, CO, USA) first intensive care unit (ICU), I saw my first case of what turned out to be acute respiratory distress syndrome. At the time, I was puzzled about a number of symptoms that I hadn't previously recognised. If the patient had been admitted in the time before intensive care wards and respirators, they would have died without the attending physician knowing why and how. This particular patient was one of the first admitted to a new ICU set up by Dr Thomas Petty and myself. We had just finished residencies in medicine and surgery, respectively, and had lobbied for and obtained the necessary funds and space. We had one primitive respirator to use on patients with respiratory failure. The patient was admitted as a surgical trauma patient to my service, with multiple fractures after an automobile accident. Despite major blood loss, he soon stabilised and had a normal chest x-ray despite a few broken ribs. Over the course of 48 h, he developed respiratory distress and was intubated and put on the respirator, which was run from the hospital wall oxygen supply on a pressure-controlled basis. Despite 100% oxygen and maximum pressure, his condition worsened and fluffy infiltrates appeared in both lungs. He proceeded to die of respiratory failure. The rapid onset fluffy infiltrates, increasing pressures that required us to ventilate the patient, and falling oxygenation puzzled me as I had not seen this in patients before. Over the following months I saw a couple of similar cases, and went to the library to see what I could find. At the time there was only one reported case of congestive atelectasis following trauma. I consulted with our pathologists and infectious disease specialists and we did find hyaline membranes and alveolar haemorrhage in our patient. Some months later I had borrowed an Engstrom respirator from our anaesthesia department. It was a volume-controlled device and superior to anything we had on our ICU. When a 12-year-old boy presented in the emergency room on a Saturday morning, I was called to come down as he was in severe respiratory distress and coughing up blood after being backed over by a pickup truck. I did an immediate tracheostomy and transferred him to the paediatric care unit. Tom Petty had recently obtained a new blood gas analyser and I learned how to operate it. I spent the morning running up and down four flights of stairs and watching the patient's arterial blood gas deteriorate as we suctioned more and more blood from his tracheotomy. He was attached to the new respirator and I noticed, off to the right hand side, a red dial labelled “end expiratory pressure”. I wondered if this might help and dialled in a moderate amount. Within minutes the bleeding slowed and then stopped and the arterial blood gas returned to normal limits. Over the next few days he was weaned from the respirator and made a complete recovery. While this was obviously a case of traumatic pulmonary contusion, it was still a case of severe respiratory distress and positive end expiratory pressure was clearly an effective treatment. Within 2 years we had collected 12 cases, which included three patients treated with positive end expiratory pressure. Getting the paper published was a struggle, and after four rejections from US-based journals, we submitted it to The Lancet, which not only accepted the paper, but published it on the basis of in-house review within 6 weeks. For the first description of ARDS see Lancet 1967; 290: 319–23 For the first description of ARDS see Lancet 1967; 290: 319–23