This study proposes to learn about the early biological and medical features of autism. Our reasoning for doing this is as follows. We observe that autism starts in infancy or early childhood. We observe that children who are diagnosed early and receive intensive early intervention do better. We also observe that many children with ASD have medical problems, such as gastrointestinal problems, or immune problems (such as frequent infections or allergies). But researchers and clinicians do not at present look for medical early warning signs of autism. If we could identify early medical and laboratory signs of ASD or high ASD risk, we might be able to reduce the severity of ASD or even prevent it entirely. But in order for this to be possible, we need to understand the earliest medical changes that occur in ASD, even before the autistic behaviors begin. Because autism has been considered a behavioral disorder, researchers have looked for early behavioral signs of the onset of autism. To do this they have studied infants who are at high risk for autism because they have an older sibling who has already been diagnosed with this condition. The goal of research into early behaviors predicting autism is to get early intensive behavioral intervention started sooner. Once we expand our view of autism to include its biomedical and metabolic features, then we can realize the importance of looking for early biomedical and metabolic signs of autism. The goal of research into early biomedical and metabolic signs of autism is to know enough to start early medical support and treatment that may improve the well-being of the child and head off some of the difficulties that child may face if left untreated. To bring us closer to the time when we understand the biomedical features of autism as much as the behavioral features, in this research project we will add biomedical approaches to the study of infants at risk for autism. We will do medical exams, measure electrical activity in the brain (through EEG), and study substances in blood, urine, and saliva to understand the chemistry and metabolism of autism. In this project we are testing the idea that the biological features of autism may come on line before the behavioral features do, and may allow us to help the child earlier than if we wait for the behavioral features. We are also operating on the belief that the biomedical and biological problems of young children at risk for autism may be a big part of the cause of autism. Even if these factors do not cause the autism--if the autism is caused independently of these medical problems--then the medical problems may still make the quality of life worse. We are therefore choosing measurements to focus on medical and metabolic problems that may closely relate to the underlying reasons why the child is having trouble with brain functioning. For example, we will study immune system functioning and see whether it changes in relation to brain or behavior functioning. And we will measure cells in the blood for signs of oxidative stress, which is a change in metabolism that happens in the cells of our bodies when we are exposed to environmental or emotional stressors. Inflammation and oxidative stress have been measured in children and adults with autism, but they have never been studied by researchers by measuring children who don't have autism yet but are at high risk of developing it. We expect that the medical and metabolic problems that researchers have been finding in older children will start to appear early in the life of a child with autism. We think that when these problems appear, they interfere with the normal functioning of the brain and, possibly, also (especially if these problems persist for a while) with the way the brain develops. We will measure this by tracking the development of medical problems at the same time as we track the development of brain signaling using EEG measures. If we see that these changes are related to each other, it will offer more ways to catch these problems early, and this may give us opportunities to correct them through treatment, and particularly through medical treatment, before they contribute to the child having more severe problems. Even if the problems don't happen for the same reasons, the body problems may be good early warning signals. Our goal therefore is to see whether we can find early biomedical predictors of autism. This study will lay the foundation for a systematic medical evaluation of every infant at risk for autism or showing signs of autism. Moreover, it will help us expand what we consider to be risk factors for autism to include problems, such as recurrent infection or early gastrointestinal problems, that may either signal that the child is vulnerable, or weaken the child and make him or her more vulnerable to later challenges--or both. If the patterns we predict turn out to be the patterns we actually find, we may be able to move rapidly to use the information to suggest that children who show warning signs on tests such as the ones we will use should get special attention and help so we can get an early start on preventing deterioration and keeping them doing well. This study will also help other researchers know where to start in investigating the early onset of autism and in looking for medical treatments that may help.