I guess that you are interested in the Pakistan area. I found a list of butterfly species at Wikipedia: http://en.wikipedia.org/wiki/List_of_butterflies_of_Pakistan.
May be some of the contacts or references quoted at the end are useful...
If it was for the butterflies in a strict sense (Papilionoidea) there is also the possibility of the books by D'Abrera, specifically the volumes on the Oriental Region. However they are very expensive. I regret not having experience on that area...
The book of P. Smetacek is very well illustrated. You can try and http://www.ucl.ac.uk/taxome/jim/lepinf.htm, but it will be good if you point the families of Lepidoptera, you will work.
Hard fact is that none of the available literature on butterflies really provide a taxonomic key based on diagnostic characters in the true sense of classification and taxonomy. However, some of the literature including the one by P. Smetacek are very helpful in easy identification of species of butterflies of certain geographical areas based on excellent pictures showing differentiating characters. Such pictorial identification guides are good for beginners and also for naturalists and biologists doing biodiversity studies, but a true taxonomist need much more than this. This is particularly so after deep changes witnessed in the classification at subfamily, tribe and sub-tribe levels due to molecular studies of a large number of butterfly taxa that do not exactly match with morphological characterization. A good species is to be judged, whether by morphology or by gene studies or both, in terms of its being a distinct and unique evolutionary unit. A sincere effort is under way to develop taxonomic key using wing venation as the basic character which has evolved more rapidly than any other characters at the species level. Morpho-types within a species are not included in it. That requires a different approach.
Please check this site http://butterfliesofamerica.com/ the most complet web site of butterflies from North, Central and South America). It contains over 160,500 full color images and list over 8,300 species. It is useful for identification at species level.
The suggested website is an excellent treasure of images of rich butterfly fauna of America. It should not be difficult to identify a species or subspecies based on pictures provided. However, the basic question of making a diagnostic key to the identification of butterflies remain unanswered! Morphological characterization is required to define a taxa. Pictorial identification based on color pattern of spots and scales on wings often overlaps and variations, even minor, are encountered from region to region. Can there be a relative constancy of characterization at species level which will avoid overwhelming dependence on external appearance of wing characters? A possible answer may lie in studying the wing venation pattern which is likely to be more constant than any other characters. Color pattern and spot characteristics may be used as supplementary characters to define its regional range. Molecular characterization is no alternative to present method of pictorial guides.