no maximum or minimum, it depends on many factors such as, research infrastructure, field, co-authors and so on

Peer-reviewing is a part of influence of the scientist to the current status of his/her discipline. Problem lies rather in the quality of peer-reviews and journals, not in number of journals. In some disciplines several proposals to be a reviewer a week is nothing strange. But if it is possible and useful? And increases scientific level of the reviewer?

For many, it is more of psychological than of scientific standing. Numbers never matter, at least in this regard. Again, the question returns another question - Though journals follow PEER REVIEWING, why do reviews differ? It boils down to the individual traits of the reviewer. If so, do the journals depend on individuals? or Individuals make the journals?

I write as Senior Editor of Archives of Disease in Childhood.

Emilia's point is well taken. Many journals provide end-of-year certificates for reviewers that can be used in academic appraisal, so there should be some pay off for the reviewer beyond the obvious fact that reviewing should provide an element of continuing professional education.

Muthuvairavasamy makes a different point, but an important one. The peer review process is not perfect: it is just that most other alternatives are worse. One of the reasons for asking for more than one review of a paper is that everyone comes with a slightly different angle, and tends to notice different strengths and weaknesses in a paper. So much is reasonably self-evident.

There is also a tendency to get the best reviews from more junior researchers: those who are very senior sometimes see fit to give nothing more than a one-liner, if they respond at all; most researchers and readers are less aware of this. But the most crucial point is that reviewers do not determine whether a paper is published. What they do is to assist the Editor who makes the decision about a paper, based on many other considerations, for example, does it fit with the journal's aim, is it interesting, is it novel, and many other factors. So while a paper with poor reviews is very unlikely to be accepted, one that reviews well may not necessarily get accepted when these other factors are considered.

In the end any biomedical journal stands or falls on its content, and reviewers are part of the complicated process of trying to ensure that the content is of high quality.

I realise ethos comment is somewhat juxtapose to the answers to date, but I am trying to remain true to the original question. A colleague once told me that it seems right to review as many manuscripts as papers that you publish yourself each year. I once passed this comment on to another colleague of mine (an editor) and they said that since it usually takes about three reviewers to review each manuscript that one should review three times as many manuscripts as manuscripts they themselves submit in a year.

There is no specific limit. Normally most of the journals will limit it to 2 to 3 per year. But you will be reviewing for a number of journals. So the actual number may be much larger depending upon your particular discipline. You will always have the option of rejecting to review based on your available time.

There is an easy math on it to make it work in a balanced way: if you publish one article in one year that during the reviewing process received 2 reviews, this would be the minimum number of reviews you may do in one year as equivalent retribution to the work of the reviewers of your published manuscript. This has a lot more details and aspects (depends of the journal, the type of manuscript, etc), but keeping this rough idea, we could guess an approximate number on how many minimum papers to review per year to be reciprocal.

Peer- reviewing is a time consuming process. The number of papers one should review per year depends on his/her load that year. Research projects, committees, teaching, your own publications, your schedule  might prevent you from committing to review many papers in a given year.

Maybe it is time for the big houses to start paying, maybe funding research or similar. "As of 2015, the academic publishing market that Elsevier leads has an annual revenue of $25.2 billion. According to its 2013 financials Elsevier had a higher percentage of profit than Apple, Inc." (Jason Schmitt)

https://medium.com/@jasonschmitt/can-t-disrupt-this-elsevier-and-the-25-2-billion-dollar-a-year-academic-publishing-business-aa3b9618d40a#.q9y71i4d7

no maximum or minimum, it depends on many factors such as, research infrastructure, field, co-authors and so on

I follow answers

best regards

The reviewing process is a time consuming

it depends upon many factors like relevant topic, your available time, your experience in such area

Depends on individual researcher priority and interest. Nevertheless from my experience there is symbiotic relationship between review and resesrch. So the more one does reviews the resesrch skills and competencies get honed. On an average I feel one review per month is fine as it helps in being updated and provides deeper reading opportunities. ..

no limitation if you have a time

I prefer to use the one at a time approach. One may accept as many as possible provided one is not busy but if one's schedule is tight, you may consider declining or notifying the editor.

One at a time is preferred for thorough assessment.

Dear it depends on the time, area of research interests, schedule in the office and ability to meet up at stipulated time.

As much as you can

Peer review is useful thing, but if you see that you diminish reading by your own choice, and instead permanently reading because of peer review this is sign that there is too much of reviewing. I try to make no more than 1 peer review/month (if speaking about original submissions).

I think you can review as many as you like as long as it's quality. It is better to review less with quality than many without quality.