Afrit is a famous name in cryptic crosswords, partly from a description of the setter's duty to the solver in this book's introduction - summarised as "You may not mean what you say, but you must say what you mean." Rather surprisingly, much of that introduction is advice to solvers. When you tackle the puzzles you'll see that Afrit's practical implementation of fairness was not as strict as that of current "Ximenean" setters who are seen as influenced by both Afrit, and Ximenes in his 1966 "Art of the Crossword" book.
You will find "forbidden practices" if you compare the puzzles to current broadsheet cryptics. (Examples are from Puzzle 1 in the book). First, indirect anagrams - "Upright attitude of a Mediterranean island (5)" = ERECT from 'Crete', with 'attitude' presumably intended as the anagram indicator. Second, partial wordplay - "Soft drink introduced by the Wild Man of Borneo (9)" = ORANGEADE as it starts with Orang (Utang) - the EADE part is not indicated. Clues are often wordy - "The man to see there's been no foul play: he'll have nothing done in a corner (7)" = CORONER - modern setters would use about 8 words. There are also a few clues that would count as racist or sexist by modern standards.
For the puzzles, you need to get used to an older style, but readers of one-puzzle-a-year anniversary collections for the broadsheets will know what to expect. In the 25-30 of the 40 puzzles I've now tried, there were a few words I didn't understand, but nothing so dated or obscure that it caused a serious problem. I gave up a few clues short of about 3 of the puzzles, and had the odd mistake in maybe 5 others.
The grids are different to present-day ones. Word-lengths in most are only from 5 to 9 letters. There are also apparent 2-letter words in most, but these are not clued. Checking is up to standard - at least half the letters in every word. The last 5 puzzles have bizarre hybrid grids using both bars and blocks.
Two puzzles are novelties - a "narrative" where clues are missing words in a passage (not too hard), and one called "True lovers knot" where four names of brides and grooms intersect. These are unclued and with most of the other answers filled in, I can't see who they are, so my guess is that this is the hardest puzzle in the book.
The standard of production is good - sewn binding, good paper, a few misprints though nothing really important. As well as the 40 puzzles, solutions, 3 pages of Afrit's Introduction and a short preface from Ximenes, we get photos and a brief note on Afrit, including an apparently accidental joke - "He was a demon at croquet" (afrit = "an evil demon in Arabian mythology"). Solutions are not explained, but very few need explaining.