The results of a retrospective study involving 622 twin-pairs born over an 18-month period among 17,726 births at the University of Ilorin Teaching Hospital, Ilorin, Nigeria, are presented with particular reference to four variables: birthweight, presentation, parity, and intertwin delivery time interval. The twinning incidence was 35.1 per 1000. Monozygous and dizygous rates are 7 and 28 per 1000, respectively. Overall perinatal mortality (PNM) was 15.5%. Mortality was higher in second than in first twin (19.5% vs. 11.6%), and consistently higher when divided into birthweight groups. Corrected PNM increased with breech presentations: 16.3% in breech:breech compared with 3.9% in vertex:vertex presentations. The twinning rate increased with parity; PNM is low in parity 1, of little variation in birth-ranks 2-5, and high in para 6 and above. Delivery of the second twin within 15 min seems optimal, giving a corrected PNM 3.6% in contrast to rates of 10.1%, 14.0% and 19.1%, respectively when delivery occurred between 16 and 30, 31 and 60 and greater than 60 min, respectively. Prevention of preterm delivery, increased use of cesarean section delivery for malpresentation, active management of delivery of second twin within an optimal time of 15 min, and family planning are suggested in order to decrease twin PNM. :Results of a retrospective study involving 622 pairs of twins born over an 18-month period among 17,726 births at the University of Ilorin Teaching Hospital in Ilorin, Nigeria, are presented with particular reference to 4 variables: birthweight, presentation, parity, and intertwin delivery time interval. The twinning incidence was 35.1/1000. Monozygous and dizygous rates are 7 and 28/1000, respectively. Overall perinatal mortality (PNM) was 15.5%. Mortality was higher in 2nd than in 1st twin (19.5% vs. 11.6%), and consistently higher when divided into birthweight groups. Corrected PNM increased with breech presentations: 16.3% in breech:breech compared with 3.9% in vertex:vertex presentations. The twinning rate increased with parity; PNM is low in parity 1, of little variation in birth-ranks 2-5, and high in para 6 and above. Delivery of the 2nd twin within 15 minutes seems optimal, giving a corrected PNM of 3.6% in contrast to rates of 10.1%, 14.0% and 19.1%, respectively when delivery occurred between 16 and 30, 31 and 60, and greater than 60 minutes. Prevention of pre-term delivery, increased use of cesarian section delivery for malpresentation, active management of delivery of 2nd twin within an optimal time of 15 minutes and family planning are suggested in order to decrease twin PNM.