RATES OF USE COMPARISONS ACROSS STUDIES ARE DIFFICULT FOR SEVERAL REASONS. FIRST, TWO DIFFERENT SAMPLING APPROACHES ARE USUALLY TAKEN: ONE SAMPLES LARGE NUMBERS FROM POPULATIONS DEFINED BY SCHOOL OR GRADE-LEVEL RANGE, AND THE OTHER SAMPLES SMALLER NUMBERS FROM HOUSEHOLDS OR CENSUS TRACTS AND DEFINES ITS POPULATION IN TERMS OF AGE. SECOND, DIFFERENT DEPENDENT VARIABLES ARE MEASURED. SOME SURVEYS INCLUDE AN ITEM-BY-ITEM INVENTORY OF A WIDE RANGE OF SUBSTANCES, WHILE OTHERS ARE LESS INCLUSIVE. FURTHERMORE, DIFFERING MEASURES OF FREQUENCY AND QUANTITY OF CURRENT AND PAST USE ARE UTILIZED, AND THE ONLY MEASURE THAT IS CONSISTENTLY APPLIED QUESTIONS WHETHER A PERSON EVER USED THE SUBSTANCE IN QUESTION. THIRD, THE RELIABILITY AND VALIDITY OF THE SELF-REPORTED DATA COLLECTION AND THE INSTRUMENT EMPLOYED MAY HAVE IMPORT; HOWEVER, THE RESEARCHER'S CREDIBILITY AND THE RESPONDENT'S FAITH IN THE CONFIDENTIALITY OR ANONYMITY OF RESPONSES APEAR TO BE MOST SIGNIFICANT. RESEARCH METHODOLOGY IS SOUND ENOUGH TO SUPPORT SOME REASONABLE CONCLUSIONS. ADOLESCENT DRINKING HAS INCREASED SINCE THE 1960'S, AND ILLICIT DRUG USE HAS SPREAD FROM THE COLLEGES TO THE SECONDARY SCHOOLS. ALTHOUGH EXPERIMENTATION WITH PSYCHEDELICS HAS LEVELED OFF AND MAY BE DROPPING, MARIJUANA SMOKING AMONG HIGH SCHOOL STUDENTS INCREASED FROM 15 PERCENT IN 1967 TO 48 PERCENT IN 1974. ESTIMATES OF HEROIN USE RANGE BETWEEN LESS THAN 1 PERCENT TO 5 PERCENT, FOR THOSE HAVING TRIED IT AT LEAST ONCE. THE PROBABILITY OF USING HARD DRUGS INCREASES WITH EACH ADDITIONAL DRUG USED; HOWEVER, ONLY A SMALL PROPORTION OF ADOLESCENTS MAKES THIS MOVE. A TABLE, FREQUENT STATISTICS, FOOTNOTES, AND APPROXIMATELY 45 REFERENCES ARE INCLUDED.