i think texture is the best criteria...

I prefer mineral composition, in consequence color...

Texture - to determine which type of rock you have, then mineral assemblage, then make a thin section to accurately determine the present minerals and their modal abundance.

During the field work, outcrop for igneous and metamorphic is totally different from the sedimentary rocks. You can differentiate in the hand specimen by grain size, colour and structure of rock types. Technically you can differentiate by texture seen under thin section, IUGS classification of major and trace elements, EPMA studies etc.

Thanks to all. If possible Share Any link or  papers in this criteria with specific photographs for pattern, mineral grain, colour and thin section analysis, would be of great work. I mind all your suggestions and search for the same.

http://blog.sina.com.cn/s/blog_69fb738a01016tmb.html

Here is some pictures but the language is Chinese ... you can see the pictures 

I would first define:

1. Texture: matrix-supported or grain-supported, massive (homogenous), glassy, porous, etc.?

2. Components: crystals, pebbles/grains (rounded or angular), fossils, biogenic components?

4. Indications for metamorphism: deformation, reactive fringes etc.

>> based on these features, decide whether it is a clastic sediment (e.g. sandstone), chemical sediment (mostly limestone), igneous plutonic (e.g. granite), igneous volcanic (e.g. basalt), metamorphic rock (can have a sedimentary or igneous parent) etc.

I find the accurate description in the first step important, because often it is not obvious from the beginning, what rock type you are dealing with.

cheers,

yuki.

Texture would be the first sign, but after that you would need a microscopic examination on the thin sections.

Nice Mr. Yuki. Very elaborate in your answer. Most needful for me, thanks. Send me some links if possible for specimen photographs in your view. As per Borislav statement I mind the  thin section analysis.

In my eyes, thin sections are only useful if you are interested in the mineralogy (i.e. mineral, or micro fossil content) of your specimens. It is a lot of work, though. For a rough classification, a hammer and a lens is enough. And some diluted hydrochloric acid, to estimate the carbonate content. As an alternative to thin sections, sawing the rock specimen and polishing the surface often reveals nice details of the texture and is much faster than preparing a real thin section.

What is actually you application? Mapping? And do you ave access to a lab?

At the moment I don't have pictures available, sorry.

cheers,y

Yes Yuki. My area is mineral and metal mapping analysis of rocks and subsequent correlations. I take your suggestions. thank you.

Use the IUGS rules. Unfortunately these are available for igneous and metamorphic rocks only.

For igneous rocks check this:

http://books.google.it/books/about/Igneous_Rocks_A_Classification_and_Gloss.html?id=u2tVu6Sbc4kC&redir_esc=y

For metamorphic rocks check this:

http://www.cambridge.org/ve/academic/subjects/earth-and-environmental-science/mineralogy-petrology-and-volcanology/metamorphic-rocks-classification-and-glossary-terms-recommendations-international-union-geological-sciences-subcommission-systematics-metamorphic-rocks

michele

Thank you ,Michele. Any link for download, if possible!

You have to buy them. It is illegal to download books with copyright, sorry.

Yes, it should be. Michele, I thought of papers. Anyway thanks a lot.

Igneous ,sedimentary and metamorphic rocks can be differentiated at first sight on the basis of their megascopic characters. For example igneous rocks are coarse-grained (if plutonic) or fine-grained (volcanic) with sutured contact ( grain to gran contact ), sedimentary rocks show bedding plane and contain cement and matrix  and metamorphic rocks contain schistose, granulose and gneissose structure which can be identified megascopically. After collecting the sample you do the microscopic study and use IUGC rule for the classification as suggested by distinguished researcher.

Thank you Dr. Dwivedi. The parameters you quoted will be very much useful to me.