We counted birds monthly from October through April of 1999–2000 and 2000–2001 on regularly grazed and rested (since 1973) semiarid grassland of central New Mexico. Horned larks (Eremophila alpestris) and chestnut-collared longspurs (Calcarius ornatus) accounted for 66% and 10% of all birds detected, respectively. We examined variation in counts of these species relative to grazing history, site, and vegetation characteristics. We used an information-theoretic approach to evaluate a set of 11 candidate models to determine which best described the variation in abundance of horned larks and chestnut-collared longspurs. There was little agreement between models for the 2 species or the 2 years. While grazing was a factor in the most likely models of abundance in the 1999–2000 sampling season for both bird species, this factor was not an important variable in explaining abundance in 2000–2001. These results suggest that abundance of horned larks and chestnut-collared longspurs are not as closely tied to grazing use during the non-breeding season in the semiarid grasslands of the northern Chihuahuan Desert, which contrasts with literature from breeding sites in the shortgrass prairie.