In decerebrate cats, we asked whether endogenous glutamate in the region of the retrotrapezoid nucleus (RTN) was involved in the control of CO2-sensitive phrenic and phrenic-related sympathetic output and, if so, which type of glutamate receptor was predominant. We made unilateral 10-nl injections into the RTN of the nonspecific glutamate receptor antagonist kynurenic acid (100 and 250 mM), the N-methyl-D-aspartic acid (NMDA) receptor antagonist 2-amino-5-phosphonopentanoic acid (AP5; 1 and 10 mM), the non-NMDA receptor antagonist 6-cyano-7-nitroquinoxaline-2, 3-dione (CNQX; 1 and 10 mM), and the inactive kynurenic acid analogue xanthurenic acid (100 mM). Each antagonist resulted in a significant dose-dependent decrease in the amplitude of the integrated phrenic nerve signal (PNA) over 30 min (CNQX > AP5 > kynurenic acid). The duration of the phrenic cycle was also decreased because of a shortening of expiratory time (CNQX > kynurenic acid > AP5). All three antagonists significantly decreased the initial slope of the PNA response to increased CO2 by 70–80% with no clear distinction in efficacy. The amplitude of the respiratory-related integrated cervical sympathetic nerve signal (SNA) was significantly decreased after kynurenic acid and CNQX but not AP5. In each case, the decrease in respiratory-related SNA accompanied a decrease in PNA and, at high levels of CO2, the decrease in respiratory-related SNA was greater than that of PNA. Endogenous glutaminergic input to neurons in the RTN via both NMDA and non-NMDA receptors is involved in the control of eucapneic PNA and timing, PNA-related SNA, and the response to increased CO2.