HR. J, in r NAT. HIST., 13: 2001 



Tabic 3. Factorial increase in length of hind-femora at each moult. 



Five-stage lem 



ales, Site A, 



1969 70 









Moult 



1st 



2nd 



3rd 



4th 



Final 



Number 



13 



26 



48 



37 



32 



Range 



1.37 1.50 



1.26 1.44 



1.26 1.43 



1.23 1.37 



1.20 1.32 



Mcan + 1 sem 



1.43 ±0.01 



1. 38 + 0.01 



1.36 + 0.005 



1.33 + 0.005 



1.25 + 0.0H-- 



Four-stage females, Site A, 



1969 70, Site B, 



1973, 1975 







Moult 



1st 



2nd 



3rd 



Final 





Number 



2 



6 



II 



7 





Range 



1.44, 1.52 



1.45 1.54 



1.34 1.44 



1.23-1.30 





Mcan± 1 sem 



1.48 



1.49 + 0.01 



1.40 + 0.01 



1.27 + 0.01 





Males, Site A, 



1969 











Moult 



1st 



2nd 



3rd 



Final 





Number 



10 



15 



20 



12 





Range 



1.37-1.54 



1.38-1.47 



1.26-1.40 



1.20 1.27 





Mean + 1 sem 



1.46 + 0.02 



1.42 + 0.01 



1.35 + 0.01 



1.25 + 0.005 





stages is in accord with Dyar's rule (1890); the rather sudden decline for the final 

 moult parallels that found for the migratory locust by Duarte (1938). The slightly 

 larger values at each moult for the females having four nymphal stages as compared 

 with those having five reflects the out-of-step growth patterns of the two types. 



Discussion 



It has been shown that the majority of female C. brunneus studied at three sites in 

 the Croydon area had five nymphal stages, while a small proportion from two of 

 these passed through only four. The first intimation of a five-stage nymphal history 

 came from the developmental pattern seen at Aberystwyth, when the possible 

 occurrence of the shorter sequence was not foreseen. Evidence for the presence there 

 of both types is however provided by seven pinned females, collected as adults, that 

 remain from this period. Antennal segmentation of six confirms a history of five 

 nymphal stages, while for the seventh it is indicative of a development completed in 

 only four. A mainly five-stage nymphal sequence, with a small contribution from 

 four-stage individuals, has also been reported by Hassall & Grayson (1987) for 

 female C. brunneus at two sites in East Anglia. This pattern of development has 

 therefore occurred in Britain across a wide geographical range. 



Variation in instar number is well recognised in non-British Acrididae and. For 

 some locust species, has been associated with phase variation (Uvarov. 1966). 

 Possibly because locusts have been much studied in the swarming phase, the lower 

 number of instars typical of this has been regarded as the norm, the solitary 

 individuals being credited with an extra stage. Historically the term "extra** has often 

 been applied to one particular instar, pictured as an interpolation between two 

 consecutive stages of the shorter sequence and duplicating, in a larger version, that 

 preceding it. Nomenclature has reflected this, so that a third instar considered as the 

 extra has been numbered Ha. This is misleading in implying, probably incorrectly, 

 that instar II is fully equivalent in both sequences. In the examples of female 

 C. brunneus encountered here it is clear that the nymphal stages, whether four or five, 

 follow a regular progression, with the growth of the hind-femora in reasonable 

 agreement with Dyar's rule, a continuous development of the external genitalia and a 



