The Essex Naturalist 57 However, north of the Thames in East Anglia it is seldom recorded. The very similar Brandt's bat (Myotis brandtii), which has still to be found in Essex, has been recorded once in Dartford and at a few other sites in Kent, (many times at Westerham). At the end of the 19th century there had been just two records of the Serotine in Essex. This is a large bat, with a preference for buildings, that emerges soon after dusk, features which, combined with its often low flight, would make it vulnerable to Victorian collectors. That there were so few records, suggests that this bat was indeed rare and Laver writes that the two Essex records "appear to be the most northerly ones recorded for the species." The present status indicates a range expansion during the last 100 years, although contemporary changes in the countryside would appear to be reducing the population of this bat. The first Essex records of Leisler's bat (then Hairy-armed bat) occurred early in the 20th century, with few reports before the present survey. It was also rarely recorded in other south-eastern counties at this time. It is uncertain whether this species was always present but confused with the Noctule; however it is now more frequently recorded in Essex than in neighbouring counties. In Kent the species is rare although a large colony of up to 164 animals was discovered in north-west Kent in 1989 (about 12km from the Aveley roost) and one breeding roost is known in Hertfordshire. This species is occasionally recorded in Greater London but is rarely found in Suffolk other than during bat box surveys in Thetford Chase. With comparable observer activity in all these areas it would seem that Essex represents a stronghold for the Leisler's bat. The Nathusius's Pipistrelle was first identified in Great Britain in 1969 (Stebbings, 1970), 21 records were summarised by Speakman et al. (1991), and it has been recorded in most years since 1990. It closely resembles the Pipistrelle and, if occurring in Laver's period, would probably have been identified as the commoner species. The Barbastelle was described as fairly common by Laver, yet has been rarely recorded since. Recent records come most frequently from Norfolk where it is found in several hibernation sites as well as a recently discovered breeding roost in a barn. It is also occasionally found in Suffolk, where again, a potential roost was found in a listed barn. It is possible that this woodland species is undetected in Essex and may still be present in small numbers in suitable habitat. Acknowledgements I am grateful to those people who have sent in records, particularly John Cox and Andrew Cook. In addition, bat counts at the Grays deneholes have been facilitated by members of the Kent Underground Research Group and woodland bunkers have been adapted by Simon Leatherdale. The distribution maps were produced using DMAP, a mapping programme devised by Dr Alan Morton. Finally, I would like to thank Phil Richardson, Dr David Corke and an anonymous referee for their comments on the drafts of this paper.