The ability of non-respiratory gases, such as nitric oxide (NO), carbon monoxide (CO) and hydrogen sulfide (H2S) to function as signaling moieties (gasotransmitters) has only recently been appreciated, due in part to their ephemeral existence in biological tissues. These gases dominated the prebiotic Earth and were major contributors to the origin of life and eukaryotes, but they disappeared as ambient oxygen levels rose and their importance in metabolism declined. Yet cells retained many of the metabolic pathways and now use them for signaling. This lecture provides an overview of these gases from the perspective of Comparative Physiology. NO is synthesized from L-arginine by nitric oxide synthase (NOS) which is present in all metazoans. Three NOS isoforms have been identified in mammals, neuronal (nNOS), inducible (iNOS) and endothelial (eNOS). Both nNOS and iNOS isoforms have been identified in fish and amphibians, whereas there is no evidence for eNOS in either vertebrate. Furthermore, fish and amphibian endothelial cells lack both nNOS and iNOS and are incapable of NO synthesis. Perivascular nitrergic nerves expressing nNOS are found in many arteries and veins of both teleost (bony) fish and amphibians suggesting that they may contribute to tissue perfusion. NO may also be generated by hemoglobin (Hb)- or myoglobin (Mb)-catalyzed nitrite (NO2-) reduction, although the physiological importance of this in aquatic animals, where ambient nitrate and nitrite concentrations may be elevated and highly variable, remains to be determined. NO is vasodilatory in most vertebrates, although vasoconstrictory responses have been reported in a few primitive fish. NO is a positive inotrope in many fish hearts. NO activation of guanylyl cyclase has been shown in all vertebrates, whereas the effects of NO on ion channels remains to be examined in fish. NOS isoforms and mechanisms of action in reptiles and birds appear to be generally similar to those reported in mammals. CO is synthesized from heme by membrane-bound heme oxygenases (HO). Three isoforms have been identified in mammals, HO-1, HO-2 and HO-3. HO-1 is inducible and stimulated by various chemical and physiological stressors. HO-2 is constitutive and activated by stimuli such as glucocortids. HO-3, may be a splice variant of HO-2; its function is unclear. Stress-induced HO-1 transcription has been shown in fish and birds suggesting it is well conserved in vertebrates. Hypoxia stimulates HO-1 expression in the hypoxia-tolerant goldfish. CO (presumably from HO-2) dilates fish vessels, and it has been implicated in a variety of neuronal signaling mechanisms in amphibians, reptiles and birds. Cysteine is generally thought to be the source of most tissues H2S production via cytosolic enzymes cystathionine β-synthase (CBS) and cystathionine γ-lyase (CSE) or the sequential activity of mitochondrial cysteine aminotransferase (CAT) and 3-mercaptopyruvate sulfur transferase (3-MST). CBS and CSE produce H2S outright, whereas release of H2S from 3-mercaptopyruvate requires endogenous reducing disulfides such as thioredoxin (Trx) or dihydrolipoic acid (DHLA). H2S may also be “recovered” from thiosulfate (S2O32-) during hypoxia. Although H2S biosynthesis appears ubiquitous in vertebrate tissues, there is considerable controversy surrounding H2S concentrations in tissues and blood. Most recent measurements suggest that H2S does not exist in physiologically significant concentrations in blood or tissues under normoxic conditions and that it most likely functions as an autocrine or paracrine signal. Vascular responses to exogenous H2S are variable as it may constrict, dilate or produce multi-phasic responses. The vasoactive effects of H2S and the inverse relationship between H2S and O2 concentration in tissues has led to the hypothesis that H2S is the enigmatic O2 “sensor” that directly couples tissue PO2 to a variety of appropriate physiological responses. In the cardiovascular system this enables pulmonary and systemic blood vessels to directly match perfusion to either ventilation or metabolism, respectively, and it appears to be a mechanism for chemoreceptor transduction of blood and environmental PO2. This O2-sensing mechanism consists of a simple balance between H2S production in the cytoplasm and its PO2-dependent oxidation by the mitochondria. Additional evidence for H2S-mediated O2 sensing in nonvascular smooth muscle is found in the gastrointestinal and urinary tracts.