Rimsky-Korsakov was an inveterate reviser of his own work, and the 'final' versions of both these scores (and of the First Symphony in particular) complete an impressive developmental curve from relative ineptitude to confident expertise. As to comparative recordings, matters are relatively simple in the case of the First Symphony, where the 1884 version serves as more or less definitive (the original dates from 1861-5). However, even there, Rimsky's stylistic antecedents are fairly conspicuous, with Schumann prominent among their ranks, especially in the symphony's very opening and the trio section of the third movement. It's an engaging piece, classical in design and reminiscent now of Glinka, now of Balakirev, but full of specifically Rimskian tensions, especially in the first two movements. Svetlanov's latest recording is a fresh, relatively easygoing affair, less assertive than his remarkably similar 1983 and 1987 versions (especially in the first movement, which has now broadened considerably), but far better recorded and marginally more subtle than either. Of rival versions, Kitaienko's is good rather than outstanding and Jarvi's, all bluster and bustle, less incisive than Svetlanov's but very enjoyable none the less.
In the case of Antar, textual complications are both rife and hopelessly confusing. Rimsky completed his very first version of the score in 1868, then revised it in 1875 and again in 1897. Until quite recently, conductors have tended to favour a 1903 re-working of the 1875 version which, according to musicologist Richard Taruskin, does not represent Rimsky's last wishes (which involved altering the second movement's basic key from C sharp minor to D minor) but was actually a compromise effected for the sake of easier engraving. Of the comparisons sampled, only Hermann Scherchen (LSO, 3/89—nla) appears to use the 1903 version, whereas both currently available Svetlanov recordings (including the Hyperion, wrongly attributed to 1897) utilize either the 1875 edition or a variation on it. In Svetlanov's first (1978) Melodiya recording (now deleted), the bass drum tremolando is missing from the opening of the second movement—a feature that all the others bar Monteux include, irrespective of the text they use. Jarvi, Maazel, Monteux and Kitaienko opt for the more ''thorough [1897] overhaul'' (Taruskin's phrase) that represents Rimsky's final wishes, and which sports a number of refinements—such as those we can hear at the end of the first movement and the beginning of the fourth.
To be quite honest, this whole area is so fraught with potential complications (contradictory scores, performing parts, personal amendments and so on), that any definitive analysis would take aeons to complete. As to the question of ''Which Antar?'', I'd personally opt for the marginally more cogent 1897 version (Rimsky's authenticated last thoughts). The 1875 'original' is more dramatically focused, especially as presented in Svetlanov's latest account which is grittier and more compelling than his 1989 Philharmonia recording, although the Allegro risoluto third movement is now an astonishing 1'44'' slower than it was five years earlier for Hyperion. The new version reports vivid brass, strings and percussion, whereas Hyperion score by capturing more of Rimsky's important woodwind detail. However, if 1875 is your final choice, then this new CD is definitely the one to go for: it's a strong, powerfully argued performance, nicely pointed in the first movement's more delicate passages and generally very well recorded. Then again, if 1897 is your preference, Pierre Monteux's 1946 San Francisco recording is possibly the most exciting of all, with Jarvi a sonically superior 'basic' recommendation. Other contenders have their virtues—the superbly recorded Lorin Maazel, for example (and it's also at mid price)—but, taken overall, Svetlanov, Jarvi and Monteux generate the most visceral excitement and parade this most attractive score in its best light.'