DALGARD AND KRINGLEN'S CONCLUSION RESTS PRINCIPALLY ON THEIR OWN FAILURE TO FIND STATISTICALLY SIGNIFICANT DIFFERENCES IN THE CONCORDANCE RATES FOR CRIMINALITY BETWEEN MONOZYGOTIC (MZ) AND DIZYGOTIC (DZ) TWINS, EVEN THOUGH EVERY STUDY THEY CITE (INCLUDING THEIR OWN) FINDS CONCORDANCE RATES FOR CRIMINALITY HIGHER FOR MZ THAN DZ TWINS. THE AUTHORS DO NOT ATTEMPT TO MEASURE THE NUMBER, SERIOUSNESS, OR RATE OR PERSISTENCE OF OFFENSES FOR THEIR SAMPLE, BUT RATHER SORT THE SUBJECTS INTO SEVEN CATEGORIES BY OFFENSE TYPE. THEIR TREATMENT OF ENVIRONMENTAL FACTORS IS INADEQUATE, BASED SOLELY ON INFORMATION GATHERED IN INTERVIEWS. NO DETAILS ARE GIVEN ON A SYSTEM OF DATA COLLECTION USED TO REDUCE BIAS. THE AUTHORS FAULT PREVIOUS STUDIES CHARGING THAT TECHNIQUES OF SAMPLING AND DETERMINATION OF ZYGOTES ARE FAULTY, BUT THEY OFFER NO SUBSTANTIATING EVIDENCE TO PROVE ERRORS IN THESE STUDIES. IT IS MAINTAINED THAT TWIN STUDIES CANNOT CLAIM TO PROVE OR DISPROVE A CORRELATION BETWEEN CRIMINALITY AND HEREDITY UNLESS SAMPLES OF TWINS SEPARATED FROM PARENTS SHORTLY AFTER BIRTH ARE USED. SEVERAL RESEARCHERS WHO HAVE USED SUCH SAMPLES HAVE FOUND SIGNIFICANT CORRELATIONS BETWEEN HEREDITY AND FACTORS RELATED TO CRIMINALITY SUCH AS HIGH LEVELS OF NEUROTICISM AND EXTRAVERSION. REFERENCES ARE PROVIDED. FOR DALGARD AND KRINGLEN'S PAPER, SEE NCJ-36243. (DAG)