Page-layout rel = «next | prev»

Along with the attribute rel = «canonical» to specify a 'search robot' to duplicate content, it is now possible to use the HTML reference value rel = "next" and rel = "prev" to indicate the position of the current page with neighbors in the navigation box. F.x. you have article splitted on 10 pages and you use next/prev buttons in order to move from one page to another. If you want to be sure that most of 'score' from 'google spider' will go to main page red="next"/"prev" is the very good options.

Here is more details about our example

www.example.com/article?id=123page=1
www.example.com/article?id=123page=2
www.example.com/article?id=123page=3
...
www.example.com/article?id=123page=10

Now about rel="next" and rel="prev"
www.example.com/article?id=123page=1
<link rel="next" href="http://www.example.com/article?id=123&page=2" />

www.example.com/article?id=123page=2
<link rel="next" href="http://www.example.com/article?id=123&page=3" />

<link rel="prev" href="http://www.example.com/article?id=123&page=1" />

www.example.com/article?id=123page=3
<link rel="next" href="http://www.example.com/article?id=123&page=4" />

<link rel="prev" href="http://www.example.com/article?id=123&page=2" />www.example.com/article?id=123page=10
<link rel="prev" href="http://www.example.com/article?id=123&page=9" />

Few notes:

• First page should have only rel=«next». 
• All pages between 1-st and last usually should have both only rel=«next» and rel="prev".
• Last page should have olny rel=«prev». 
• Allowed to use the value rel=«previous» as an alternative to rel=”prev”. Not sure if it has sence :) 
• In case if you fail with that Google will continue to index your content with own heuristic logic.

sources: View-all in search results and Page-layout rel = «next | prev»