THE ORIGINAL SCHEME for a Hurtig encyclopedia, which was first invited and then turned down by the Canada Council, was drawn up by Morris Wolfe and me, and we were going to be its editors. I mention this at the outset in case any reader who remembers the fact suspects that it colours my response to the actual book. It doesn't, because this work is very different from what we envisaged, and much more ambitious. Our plan was for a single volume modelled on The Columbia Encyclopedia, with some utilitarian line drawings but no other illustrations. The Canadian Encyclopedia as first published in 1985 was in three volumes, lavishly illustrated.

The second edition, appearing as planned after the remarkably short interval of three years, has four volumes. The text is longer by 371 pages, the index by 265. And here is the most noticeable imptovement, as the index of the first edition was its one major flaw. I say this with no smugness; in the original scheme there was to be no index at all. As in Columbia, subjects not given articles of their own were to appear in alphabetical order in the main text, with cross?references to the articles in which they were mentioned. This was perhaps not a good idea, but the system adopted in the first edition was certainly a bad one: the headwords of articles didn't appear in the index alphabetically, but were used instead of page numbers. An instruction at its head said: "The reader should ?look for a subject first among the articles of the encyclopedia." Thus, if you wanted to know about George Bain, you pulled Volume I from the shelf and found there was no article on him. You went on to the index in Volume III, and it sent you to the article "Globe and Mail" in Volume Il.

This clumsy system has been abandoned in the second edition, which has a full index ? fuller than it need have been, perhaps. Would it ever occur to you to look up "Latin language" in an encyclopedia of Canadian things? Yet it's in the index, with references to the articles on "Bearberry," "Mineral Resources," and "Rodeo," among others. Still, it's better to have too much in an index than too little, and I've gone on about this improvement at some length in order to assure owners of the first edition that if they use it a lot the second edition is well worth acquiring for this reason alone.

Some articles have been usefully lengthened, many added, and none subtracted as far as I can discover. The first 14 pages, for example, running from "A Mari usque ad Mare" to "Administrative Law," correspond to eight pages in the first edition. Here are the additions. Under the portrait of J. J. C. Abbott, the third prime minister, is a table of such information as his parents' names, his father's occupation, where he was educated,. his religion, his occupations before he entered Parliament, his constituencies and portfolios, his wife's name, the number of children they had, and when he was knighted. There is a similar table for each prime minister; it's just this sort of utilitarian detail that I find myself using reference books for much more often than for reflective essays. I wish it had been done for other figures.

The next noticeable additions are a lengthening of the article on "Academic Freedom" and a new article, "Academie canadienne?francaise." Then comes a major new development: the replacement of a rather perfunctory two?column article on "Acadia," which started with a slightly imprecise account of the origin of the name, by a new article occupying nearly 12 columns giving the history of Acadia in considerable detail, a description of contemporary Acadia, and (the biggest section) a survey of folklore, music, painting and sculpture, theatre, cinema, and literature. That's the stuff we want.

Next, the article "Acid Rain" has been lengthened and embellished with a diagrammatic painting to show how it happens and a map to show where it happens. The rest of the additions are three short articles: one, mislabelled "ACTRA Awards," is really on ACTRA ? only the last two sentences are about the awards; an article on the early Toronto publisher Graeme Mercer Adam; and one on the contemporary singer Bryan Adams.

I choose these first 14 pages not, as you probably suspect, because they're as far as I've got, but because these additions and improvements seem to me fairly representative of the excellent encyclopedic thinking that has gone into the new edition and the range of subject?matter to which it has been applied.

Still in the letter A, I wonder whether anyone wanting to know about arthritis would reach for an encyclopedia on Canada. The article on it is good, but the ailment is the same the whole world over. However, this edition adds one on the Arthritis Society, which is a Canadian subject. I suggest that in the next edition the first article, in a shorter form, could be merged into the second.

I long wanted a new reference book on the native peoples to update and replace the useful little Handbook of Indians of Canada issued as a Parliamentary Paper in 1913 and reissued in 1971 in the Coles Canadiana Collection. Since 19,85 that has been unnecessary; the articles in this encyclopedia cover the subject with extraordinary thoroughness. Look under "Native People" (it should have been "Peoples," but never mind) and you will find 14 articles occupying 26 pages, amply illustrated, and with cross?references to many more articles throughout the four volumes. I have a publishing suggestion for Hurtig: look into the feasibility of gathering these articles into a single volume to be issued separately for those who are interested in this subject but don't want or can't afford the whole set ? or, like me, are usually in a hurry and find one volume handier than four.

As I have said, I use reference books for hard facts much more often than for reflective essays. One of the hazards of using them in this way is that one's eye is often caught by an entry of the thoughtful kind, which imperiously demands to be read. Thus, when all I wanted to know was the date of the founding of the League for Social Reconstruction, I was entrapped by the article "Leadership Convention" by John C. Courtney. It's everything an encyclopedia entry should be. While giving all the names and dates that are needed, Courtney manages in very small space to show and discuss the significance of the phenomenon. He points out that Canada is the only one of the parliamentary countries to have adapted the American nominating convention to the purpose of selecting parliamentary leaders, and that this usually results in the choice of a leader with little or no experience in the House of Commons. He concludes: "It is debatable whether the openness and freedom inherent in the leadership?convention system . . . is necessarily beneficial to the larger political system." He gives you the facts and requires you to think.

In the first edition, I thought it admirable to give the full text of the Constitution Act, 1982, but since that act is in part an amendment to the BNA Act (or the Constitution Act, 1867, as we must now call it) it seemed odd not to include it as well. It remains, after all, the basic document in the written part of the constitution. I assumed the omission would be rectified in the enlarged second edition. It isn't.

Unlike George Galt, who reviewed the first edition in these pages, I thought the illustrations were an admirable feature, especially the beautifully printed colour photographs that effectively evoked many of the places described in the text, and the small but clear maps that showed the locations of those places. They are still here, and more. For me the trite 1920s?style endpapers in the first edition were a serious blemish. And now they have been replaced by a handsome perspective map. Congratulations all round.