*COMMENT ~ Character notes. 
 
*CHARACTER NOTES 
#4. The ranges given here are mostly taken from Lucas (1900), with the
interpolation of mean values quoted from Hammond (1983). \par{}When
constructing keys for practical application, the existence of intraspecific
variation is no excuse for downweighting differences of obvious identificatory
value in favour of obscure morphological features, regardless of the supposed
taxonomic importance of the latter. See further comment under wingspan. 
#5. Measured with the front edges of the hindwings at right angles to the
body. The figures given are quoted from Hammond (1983). \par{}This character
exemplifies very well the value of simple estimates of size as primary
identificatory separators, although they are invariably ignored in favour of
obscure morphological features in printed keys. In the present case, the data
on wingspan remain a very useful primary separator in INTKEY, even
automatically allowing (as here) for variation of 20% above and below the
means given by Hammond. See Percent error, in Dallwitz, M. J., Paine, T. A.,
and Zurcher, E. J. (1995 onwards): \i{}Users guide to Intkey\i0{}. 
#6. Measurements are quoted from Hammond (1983), cross referenced with Lucas
(1900). \par{}When constructing keys for practical application, the existence
of intraspecific variation is no excuse for downweighting differences of
obvious identificatory value in favour of obscure morphological features,
regardless of the supposed taxonomic importance of the latter. See further
comment under wingspan. 
#34. Measurements are quoted from Hammond (1983), cross referenced with Lucas
(1900). \par{}When constructing keys for practical application, the existence
of intraspecific variation is no excuse for downweighting differences of
obvious identificatory value in favour of obscure morphological features,
regardless of the supposed taxonomic importance of the latter. See further
comment under wingspan. 
#46:48:50:54-57:59-60:64-70:83. Identifications of nymphs cannot be pursued
with optimism until they are sufficiently mature; i.e., until the wing sheaths
reach beyond the third abdominal segment. 
#84. It is axiomatic that no identification can be relied upon until it has
been confirmed with reference to a detailed description. \par{}\par{}Known
distributions of organisms are obviously taxonomically useful, but equally
obviously, they have to be used with caution. The very detailed data provided
by Hammond and Merritt (1985) have therefore been geographically generalized
here, mainly to render them practicable and reasonably reliable for helping
with identifications. The distributions have automatically been widened, by
recording them quite liberally under the broad regions used for this
character; and the routine advice we advocate when using INTKEY (if in doubt,
select more than one character state) remains available to users as a further
precaution. \par{}\i{}Southern Scotland\i0{}: south of the Firth of Forth.
\par{}\i{}Northern England\i0{}: including north Derbyshire, Cheshire,
Lancashire, Yorkshire, north Lincolnshire. \par{}\i{}English Midlands\i0{}:
central England, including Warwickshire, Northamptonshire, Leicestershire,
Nottinghamshire, southern Lincolnshire, Derbyshire, Staffordshire,
Worcestershire, Shropshire, Herefordshire, etc. \par{}\i{}East Anglia\i0{}:
eastern England south of The Wash, including Norfolk, parts of Cambridgeshire
and Essex. \par{}\i{}Southeast England\i0{}: including London and the Home
Counties, Kent, East Sussex. \par{}\i{}Central southern England\i0{}:
including Oxfordshire, Gloucestershire, Buckinghamshire, Hampshire, east
Dorsetshire, Wiltshire. \par{}\i{}Southwest England\i0{}: west Dorset,
Somerset, Devonshire, Cornwall. 
#86. It is axiomatic that no identification can be relied upon until it has
been confirmed with reference to a detailed description. \par{}\par{}Known
distributions of organisms are obviously very useful for identification, but
they have to be used with caution because from time to time, specimens will
encountered outside the recorded ranges of species. Precise, up-to-date
vice-county records have not been available for the present purpose, and
distributional data from the Hammond and Merritt maps (1997) have been
deliberately widened for encoding. The geographical generalization should
render the information more reliable for helping with identifications, and the
routine advice we advocate when using INTKEY (if in doubt, select more than
one character state) remains available to users as a further precaution. 
#87. It is axiomatic that no identification can be relied upon until it has
been confirmed with reference to a detailed description. \par{}\par{}Known
distributions of organisms are obviously very useful for identification, but
they have to be used with caution because from time to time, specimens will
encountered outside the recorded ranges of species. Precise, up-to-date
vice-county records have not been available for the present purpose, and
distributional data from the Hammond and Merritt maps (1997) have been
deliberately widened for encoding. The geographical generalization should
render the information more reliable for helping with identifications, and the
routine advice we advocate when using INTKEY (if in doubt, select more than
one character state) remains available to users as a further precaution. 
#88. Adults found earlier or later during the flight periods given are likely
to exhibit immature or very mature (i.e., atypical) colouration. 
 
