Were the purpose only to ascertain the _acoustic_ properties of unadulterated air, varied experiments might eliminate the errors of anomalous results. But when the process is reversed, and we deduce _effects_ from _only one_ among concurrent and conflicting causes, theory is confounded by discordant facts. Theories of _sound in purely elastic air_ might give results approximately realized in practice, if the actual pulses with which we are concerned were given by a flail; but are pregnant of error when the atmosphere is mixed with vicious vapors, and the _pulse_ is a breath of air. Then, the assumption that “_pulses of sound_” proceed _equally_ in all directions from the initial point, is simply false; and theories based upon it can only complicate the problems to be solved. Water, as well as air, is a highly elastic fluid, and, if confined and subjected to pressure, the force applied is exerted on all sides of the confined volume. But the _effect_ of a _pulse_ or blow upon a surface of large extent varies with the _direction_ of the force as well as with its _power_ and _velocity_. We have seen fish swimming near the surface killed or paralyzed by a blow upon the water immediately over them. And we have seen the blow fail of its intended effect _solely_ because it was _misdirected_. Perhaps the water in the latter case was _not perfectly elastic_! Neither is the air of churches and public halls, when their atmosphere has yielded a portion of its _oxygen_, and, in return, is charged with _carbonic acid_ and moist vapors from the breath of crowded assemblies. _Carbonic acid gas_ is heavier _by one-half_ than atmospheric air. It does not, then, always rise toward the ceiling or roof, but remains in solution with impure exhalations; or else, condensed by contact with the colder walls, descends to poison the lower air and impair its elastic force--its power of transmitting the “_pulse of sound_” to the ear. We have just come from one of our city churches, where we have had a striking example of this result. The church in question will _accommodate_ (?) about two thousand people. Twenty-five hundred may be crowded into it. At the commencement of the sermon, the preacher’s voice was distinctly audible at points fifty or sixty feet from the pulpit, in spite of reflections of sound--_air pulses_--from galleries, wooden columns, and the arched ceiling and side-walls, of lath and plaster. Before it was ended, the exhalations of the breathing crowd had so filled the lower half of the “_auditorium_” that only vowel sounds could be distinguished; and the _peroration_ seemed to consist of spasmodic utterances--scarcely sounds--of _a_, _e_, _i_, _o_, _u_. _W_ and _y_ had lost their affinity to vowels, and the rest of the alphabet were no longer _consonants_, for they were not heard at all.