We examined the relation of pre-pregnancy body size to chromosomally normal and chromosomally aberrant spontaneous abortion. Data derive from a hospital based case-control study of spontaneous abortion in the public and private facilities of three New York City hospitals. Chromosomally normal (n = 1265) and chromosomally aberrant (n = 872) spontaneous abortions (cases) were compared with 3795 women attending prenatal care before 22 weeks of gestation and delivering at 28 weeks or later (controls). Data on height and pre-pregnant weight were obtained by interview and Body Mass Index (BMI, weight/height squared) was computed. Associations were consistent across payment strata with disparate sociodemographic characteristics. BMI was similar in chromosomally normal cases and controls (pooled adjusted mean difference = 0.21 kg/m squared, 95% CI -0.06, 0.48) and reduced in chromosomally aberrant cases compared with controls (pooled adjusted mean difference = -0.29 kg/m squared, 95% CI -0.58, 0.00). These associations are unlikely to be due to reporting bias. However, they were of small magnitude and the association with chromosomally aberrant loss did not differ from that with chromosomally normal loss, nor was it specific to one, or even two, types of aberration. We therefore infer that, in relatively well nourished populations, low pre-pregnant body size does not increase the risk of either chromosomally normal or chromosomally aberrant spontaneous abortion.