 Solution

Use a documented examination of the whole body and lungs, with a lung flotation test as corroboration rather than an infallible verdict. Previous breathing, live birth and death by drowning are three different propositions. The question asks how to investigate; it does not supply a body on which to report an actual positive or negative result.

First establish the apparent age and development of the child, the condition of the remains and any evidence about birth, recovery from the water or attempted resuscitation. If the child is evidently older, independent evidence of survival after birth may settle much of the issue. In a newly born child, the forensic problem is harder. Record decomposition, injuries and any manipulation before opening the chest. Avoid inflating or unnecessarily compressing the lungs before their original state has been documented.

On opening the chest, observe the position, extent, colour and consistency of the lungs. Unexpanded newborn lungs are generally compact and do not fill the chest as expanded lungs do. Aerated regions may be more expanded, spongy and crepitant. Cut surfaces may show small air bubbles. These observations support a judgement about aeration, but disease, collapse and decomposition complicate their interpretation. Examine both sides and different regions, because a few breaths may expand only part of the lungs.

For the historical lung flotation test, place the removed thoracic organs in water, then examine the lungs separately, their lobes and representative small pieces. Record what floats and what sinks at each stage, using a vessel deep enough for free flotation. The physical principle is reduction of average density by retained gas. Testing only a whole organ can miss a small aerated region. The test belongs to the pre-1865 forensic toolkit: the contemporary encyclopedia discussed lung flotation and explicitly recognized both artificial inflation and putrefaction as alternative explanations for floating.^[ref]

There are two main classes of error. Floating can result from gas introduced by attempted artificial respiration or generated after death, so it does not by itself prove spontaneous breathing. Look for decomposition elsewhere and compare the distribution of gas in the lungs and other organs. A floating piece of liver can support the suspicion of generalized gas formation, but a sinking comparison organ does not guarantee that all pulmonary gas arose during life. Historical claims that squeezing a fragment always distinguishes breathed air from artificially introduced air were too confident; compression is not a decisive separator of the causes of aeration.^[ref]

Sinking does not establish that the child never breathed. A few ineffective breaths, limited expansion, or heavy, fluid-filled or diseased lung tissue can leave the material tested denser than water. A later prospective autopsy study reported four false-negative tests in which lungs sank despite prior life reported by medical staff. That observation demonstrates why a negative result cannot stand alone; the study's overall performance in its own material is not a universal accuracy guarantee for decomposed bodies recovered from water.^[ref]

Use the whole examination and independent evidence to reach a qualified conclusion. Feeding, healing changes or reliable observations of the living child may support survival, but they are different evidence from lung aeration. A child can show signs of life without establishing effective breathing, and an inflated lung does not establish a particular cause of death. Immersion may have occurred before or after death; neither flotation nor the body's location alone proves drowning. Where decomposition or attempted resuscitation prevents a confident distinction, report the limit rather than manufacture certainty.