A MAIN CONCERN OF THE 2 STUDIES DESCRIBED IN THIS PAPER WAS TO DEVELOP A METHOD OF COLLECTING LANGUAGE SAMPLES OF CHILDREN. IN THE BELIEF THAT LANGUAGE SHOULD BE STUDIED AS A PART OF SOCIAL INTERACTION, THE CHILDREN'S SPEECH WAS RECORDED AS IT WAS BEING USED IN FAMILAR HOME AND SCHOOL SETTINGS. IN THE FIRST STUDY (TO AVOID INTRODUCING ALIEN ELEMENTS WHICH MIGHT HAVE MODIFIED NORMAL LANGUAGE) A TRANSMITTER WITH AN INTERNAL MICROPHONE WAS CONCEALED INSIDE A 3-YEAR-OLD CHILD'S CLOTHING. HE COULD MOVE ABOUT FREELY AS HIS CONVERSATIONS WERE BROADCAST TO A BATTERY-OPERATED RADIO AND RECORDED ON A TAPE RECORDER. IN THE SECOND STUDY, NURSERY SCHOOL SUBJECTS (AGED 4), EACH WORE A SMALL POCKETBOOK CONTAINING A TRANSMITTER. A MICROPHONE WAS CLIPPED TO THE POCKETBOOK'S SHOULDER STRAP, ABOUT 6 INCHES FROM THE CHILD'S MOUTH. FIVE MINUTE SPEECH SAMPLES WERE RECORDED EVERY 20 MINUTES OVER 2 MORNINGS FOR EACH CHILD. SOUNDS WERE TRANSMITTED TO A TUNER AND TAPED. AN ADVANTAGE OF THE METHODS OF COLLECTING LANGUAGE SAMPLES USED IN THESE STUDIES IS THAT THE RECORDINGS HAVE A POINT OF VIEW, WITH THE CHILD'S SPEECH RELATED TO ANOTHER PERSON'S. TRANSCRIPTS OF THE LANGUAGE SAMPLES ARE INCLUDED. THIS PAPER WAS PRESENTED AT THE BIENNIAL MEETINGS OF THE SOCIETY FOR RESEARCH IN CHILD DEVELOPMENT (NEW YORK CITY, MARCH, 1967). (MS)