Pyroxene and maskelynite are major minerals in basaltic Martian meteorites (Shergotty, Zagami, EETA79001, and QUE94201). They are chemically zoned in both major and minor elements, offering useful information on their crystallization history. Pigeonite and augite show two distinct textural occurrences of zoning patterns. In Shergotty and Zagami, pigeonite and augite are usually present as separate grains that are zoned from Mg-rich core to Fe-rich rim, respectively. Both pigeonite and augite usually have homogeneous cores, considered to be cumulus phases. Zagami pyroxenes are not zoned as extensively as those in Shergotty, but their mineralogy is quite similar. On the other hand, pigeonite and augite in EETA79001 and QUE94201 are both present in individual composite grains. These pyroxenes are complexly zoned, and typically the cores are Mg-rich pigeonites, mantled by Mg-rich augite, and the rims are Fe-rich pigeonite. Pyroxenes in lithology A of EETA79001 are small and show irregular zoning patterns, but most of them show similar zoning patterns to lithology B of EETA79001 and QUE94201. Maskelynite compositions also correspond to differences in pyroxene zoning. Maskelynites in Shergotty and Zagami are more alkali-rich than those in EETA79001 and QUE94201. Shergotty and Zagami maskelynites apparently nucleated on pyroxene crystals and grew outward (to lower An content) into interstitial melts, whereas EETA79001 and QUE94201 maskelynites have the most An-rich portions in the centers of grains, showing normal core-to-rim zoning. FeO in the maskelynite cores is different between Shergotty and Zagami (0.5–0.6 wt%) and EETA79001 and QUE94201 (0.3–0.4 wt%). The lower Fe content of EETA79001 and QUE94201 maskelynite cores reflects earlier crystallization of plagioclase. Al zoning in pyroxenes also marks the beginning of plagioclase crystallization. Decrease of the Al/Ti ratios observed in Al–Ti plots of pyroxenes further suggests plagioclase crystallization, and the Al–Ti distribution shows a clear difference between Shergotty and Zagami on one hand and EETA79001 and QUE94201 on the other. Such mineralogical differences of EETA79001 and QUE94201 from Shergotty and Zagami can be understood by undercooling of the magmas from which they have crystallized. We believe that Shergotty and Zagami experienced only slight undercooling, resulting initially in cotectic growth of pigeonite and augite, later joined by plagioclase. On the other hand, EETA79001 and QUE94201 experienced significant undercooling of their melts, and their crystallization sequence was pigeonite, augite, plagioclase, and then Fe-rich pigeonite, with each phase crystallizing metastably alone, because the melt compositions did not follow the equilibrium phase boundaries. Lithology A of EETA79001 also experienced undercooling effects, but they are not so extensive as in lithology B, inasmuch as lithology A shows some features similar to Shergotty and Zagami. All of these inferences concerning supercooling are consistent with field emission gun scanning electron microscopy (FEG-SEM) observation of pyroxene microstructures if we consider that magmatic cooling rates are related to subsolidus cooling rates. That is, the pyroxene cores of Shergotty and Zagami have fine exsolution lamellae (a few tens to a few hundreds nm), whereas those of EETA79001 and QUE94201 do not contain such exsolution lamellae, indicating faster cooling rate even at subsolidus.