I personally use a 5 step method.

1. Carefully examine the wing pattern of all 4 wings, upper and under sides, and make comparisons with illustrations and identified specimens (if you trust the identifications).

2. Examine the external morphological characters, e.g. leg spines and “claws,” setae on the eyes, antennal pectinations and setae, abdominal tufts, structure and density of scales on wings, head, abdomen, thorax, legs, etc. Examine structure of eye lashes, thoracic crests, tympanum and others that can be seen with a low power microscope or triplet hand lens. Use the literature to compare the characters you observed with known taxa.

3. Dissect the abdomen to examine the genital parts. Use the literature to compare the characters you observed with known taxa. Literature with illustrations is most useful.

4. Submit legs to a lab for DNA analysis. This step requires some research to find a lab, and some funds to pay for the analyses.

5. Make a Wild Ass Guess. This is often a good place to start by comparing your specimen with illustrations or specimens in a curated collection. The guesses will become more discriminating as you follow steps 1 and 2 above.

At each step in this process I consult colleagues, literature (especially books with illustrations), and other collections.

Good luck.

Eric

PS:  When I say literature, I include web sites.

Eric has more or less laid down the procedure. I would like to add that in the case of Asian Noctuidae, many are not known and few have DNA profiles, so one would need to be cautious. 

When a person finds a new taxon, it is distinguished from other members of the genus or sub-family by the singular features it possesses, so there is no rule of thumb for separating species- it could be some feature on the wings, or external morphology or, in the case of cryptic species, a difference in DNA. Step carefully, especially with this family! Good luck!

Peter is absolutely correct.  When I'm in any faunal area, I first start out by searching literature and creating a list of possible species I might find.  I create a file, and when possible I add an illustration in the document.  This document then becomes my personal guide to the described species in the faunal area.  In the Noctuidae, the list is very long for any region.  It seems like a tedious process, perfect for an anal person like me, yet when I find a species to identify, my advance work makes the final task much easier.  I could not describe species without conducting this groundwork.

When one wants to identify a noctuid, some basic observations make the process mucdh easier.  When a person shows me an unknown noctuid, I first ask if that person looked at the eyes and legs.  If not, I show them how to do it.  Better to teach how to do it rather than doing it myself.

well yes, for identifying noctuid moths we should consider a mix of features however some of them are to be excluded or included, and if you ask me, some times i'm really confused to separate one species from the others, but the five-step method written by Eric is right! it's not precious to rely on one method!

Dear Masood:

The owlet moths are robust, stout-bodied insects, usually with filiform antennae. The fore wings tend to be mottled in color, often cryptic, and slightly longer and more narrowed than the hind wings. In most, the hind wings will be brightly colored, but kept hidden under the forewings when at rest. Some owlet moths have tufts on the dorsal surface of the thorax.

If you want to ID by studying wing venation details, you should note the following traits in the owlet moths you collect:

The subcosta (Sc) arises near the base of the hind wing.

The subcosta (Sc) fuses briefly with the radius near the discal cell in the hindwing

Three medio-cubital veins extend to the distal edge of the hind wing (see the wing venation attached).

The molecular Phylogenetics of the Noctuid moths was recently studied by Sahiri et. al. 2011. Please see PDF file attached

Is there any classification for the arrangement of cornuti in the aedeagus?