Louis Agassiz Shaw Jr.

Louis A. Shaw, identified in Harvard's institutional timeline as Louis Agassiz Shaw, Jr., coauthored the early tank-respirator reports with Philip Drinker.^[ref] Their apparatus applied changing pressure around the body while the head remained outside, implementing negative-pressure ventilation rather than requiring the same pressure change at the airway.

The second 1929 report, led in the published byline by Shaw, adapted the apparatus for small children and infants. It describes a metal enclosure, rubber neck collar, pump and valve arrangement, with a window for observation. Importantly, the report said the new apparatus had not yet been available for clinical use and cited laboratory tests on curarized cats. Those statements distinguish a reported experimental test from demonstrated clinical outcomes in the intended population.^[ref]

A chamber seal is part of the mechanism. An original steady leakage model lets $P$ be suction magnitude relative to room pressure, $G$ the chamber-leak conductance and $Q_p$ the pump's extraction volumetric flow rate over the modeled operating range. If leakage is linear,at steady state. With hypothetical $Q_p=5$ L/s and $G=0.5\mathrm{L}{\mathrm{s}}^{-1}(\mathrm{cm}{\mathrm{H}}_2\mathrm{O}{)}^{-1}$, suction is 10 cm H${}_2$O. Doubling the leak conductance at the same extraction rate reduces it to 5 cm H${}_2$O. These are illustrative values, not apparatus specifications from the report.

This chamber-leak sensitivity in negative-pressure ventilation separates chamber pumping from pulmonary ventilation. Pumped air through a leaking enclosure is not automatically air passing through the subject's airway. In a further ideal passive mechanical equilibrium model with total respiratory-system compliance 0.01 L/(cm H${}_2$O), those two driving pressures would produce added volumes 0.10 and 0.05 L. Measuring pump operation alone does not establish delivered lung volume.

The historical research therefore joined physiological reasoning with enclosure geometry, sealing and observation. A successful mechanical cycle tests that the apparatus changes its boundary conditions; independent physiological measurements are needed to establish what breathing assistance those changes provide.