Fig. 4.a Marie’s non-aphasic case with a destructive lesion which Marie thought had destroyed Broca’s area but actually the lesion is located in Area 46. b Marie’s case with Broca aphasia without lesion on the Broca’s area. Areas 44 and 45 are missing. c Dejerine’s case cited by Marie as a case of Broca aphasia without lesion on the Broca’s area. Area 45 of the Broca’s area is evidently affected. (a cited from [1], Band c cited from [2]).

      In his second paper [2], Marie showed an illustration of brain slice of Case Bal…, who clinically showed Broca’s aphasia (Fig. 4b). According to Marie, the lesions of this patient affected the “zone lenticulaire” and the deep white matter of the Wernicke area together with the white matter of the temporal lobe. But again, since both opercular and triangular parts of Broca’s area are missing from the illustration, we can never know whether these most important cortical areas are intact or not. In this paper [2], Marie cited the illustration of the brain slice of Bernheim’s case with Broca’s aphasia which appeared in Dejerine’s book (Fig. 4c). Marie discussed the clinicopathological similarities between these two cases, but the illustration of the Bernheim’s patient cited in Marie’s paper clearly shows that the opercular part of the inferior frontal gyrus corresponding to Area 44 is preserved but the triangular part of the same gyrus corresponding to Area 45 of the Broca’s area is totally damaged and atrophied. Marie seemed to have repeatedly made errors in identifying the anatomical site of the Broca’s area not only in his own cases, but also with the anatomo-pathological reports of other investigators [6].

      Why Marie Did Not Notice the Missing Parts?

Then the question arises. Why Pierre Marie had repeatedly made this kind of simple error to identify the cortical areas? The answer is given by one of his last pupils form the United States. Percival Bailey, the famous neurosurgeon, entered Marie’s clinic at the Salpêtrière in 1921 as foreign assistant. He wrote “Only rarely did he enter the wards and never the laboratory, for he was very sensitive to formalin and would look at fixed brains only through a window, and there dictate his description” [7]. Although the above-mentioned anatomo-pathological studies of patients with aphasia were done in the Hospice de Bicêtre, and not in the Salpêtrière, he was sure to use it to observe the already cut horizontal brain slices indirectly through a window, as Bailey described. As shown in the previous Figures 1 and 4, two most important pieces of the posterior part of the inferior frontal gyrus, opercular and triangular parts are liable to be dropped out from the entire horizontal slice at the level of Sylvian fissure because they are not connected with the rest of the brain. My speculation is that Pierre Marie, in his routine laboratory works, made his assistants to cut the brains, but they did not pay enough attention to these 2 small pieces of brain inevitably detached from the rest of the brain slice.

      Dejerine’s Error in Identifying Triangular Part

Similar error of identification of the easily detached small pieces of Broca’s area was also seen in Figure 225 of Dejerine’s “Anatomie de Centres Nerveux, Tome 1” [8]. The figure (Fig. 5) is a horizontal slice of normal brain and clearly shows two small pieces within the Sylvian fissure. Dejerine identified the posterior piece as Rolandic operculum and the anterior piece as frontal operculum, and the deep fissure between these two pieces as frontal opercular incisure, but the truth is that this deep fissure is ascending ramus of Sylvian fissure and the anterior piece which Dejerine identified as frontal operculum is actually the triangular part of the inferior frontal gyrus. Due to this misidentification of the ascending ramus of Sylvian fissure, Dejerine noted erroneously the above-mentioned anterior piece of brain as the opercular part and the cerebral gyrus situated anterior to this fissure as the triangular part (cap de la troisième circonvolution), but his anatomical identification was incorrect (Fig. 5).