Discussion

Morphological bone data obtained from ground sections affords an understanding of the degree of vascularization in dinosaur remains. This attribute has direct relation to dinosaur physiology, life cycle, growth rate, behavior, feeding specializations, sensory capability, disease, parasitism, and predation [Reference Chinsamy1,Reference Clarey3,Reference Woodward7,Reference Colquhoun10,Reference Horner13–Reference Brink20,Reference Cullen22,Reference Aureliano36]. Bone morphology can also be used to estimate the actual age of the dinosaur at death [Reference Woodward7,Reference Horner and Padian14].

Many rare dinosaur bone specimens are discovered crushed, fractured, damaged by predators, weathered, or accompanied by artifacts of taphonomic processes [Reference Currie2,Reference Yun5,Reference Horner13,Reference Schweitzer17,Reference Brochu37]. This is especially true of skulls [Reference Clarey3,Reference Brochu9,Reference Gilmore23,Reference Schmerge and Rothschild29,Reference Schmerge and Rothschild30,Reference Carr31,Reference Witmer and Ridgely32,Reference Brochu37–Reference Witmer, Endo and Frey39]. Thus, the quality and anatomic accuracy of neurovascular canal images from such specimens can be adversely impacted. In some cases, only a few or even one specimen of a given taxon exists. Additionally, sectioning of bone (especially serial sectioning) is a destructive technique and may be cost-prohibitive depending on the size and shape of the bone. Moreover, workers are simply reluctant to subject rare or exceptionally well-preserved specimens to such a process. Other problems exist, like the difficulty in rendering 3-D images from serial ground sections, and the fact that the entire bone is made unavailable for further study after serial sectioning. Further, collection, processing, and sectioning methods can introduce artifacts, rendering images that cannot be interpreted.

In recent years, a non-destructive computed tomography (CT) scanning of dinosaur bones has also been employed, resulting in remarkable images of neurovascular and other canals [Reference King6,Reference Bouabdellah8–Reference Brochu9,Reference Brochu37–Reference Kawabe and Hattori43]. This is especially notable with respect to imaging dinosaur jaws (maxillae and mandibles) [Reference Bouabdellah8,Reference Bell40,Reference Kawabe and Hattori43]. Soil matrix and metal artifacts can be identified by CT scanning [Reference Brochu37], however none of the referenced dinosaur bone CT studies have reported blood products in neurovascular canals. Soft neurovascular elements are not visualized in isolation, and only the cranial cavities are imaged in digital endocasts [Reference Cerroni42].

Bone has a large mineral component yet is amply supplied by soft tissue neurovascular bundles [Reference Bouabdellah8,Reference Sakagami and Kawabe42–Reference Kawabe and Hattori43]. It would be expected, therefore, that a rich supply of blood vessels, veins, and nerves would have been present in dinosaur bones at death. A post-mortem condition known as livor mortis occurs when gravity-created blood pools form at the lowest part of a decumbent body ([Reference Schweitzer17] pp. 64–65). Blood pooling within dinosaur bones might have occurred as well but was not studied.

Reports of blood in histological sections or decalcified soft material from dinosaur bones are rare [17,44 (Fig. 16),45]. One worker reported observations of “small red spheres … which lay in a blood vessel channel that wound through the pale, yellow hard tissue,” [Reference Schweitzer17]. Moreover, the report noted, “Hollow, transparent, flexible branching tubes … and they looked exactly like blood vessels. Suspended inside the vessels were either small, round red structures or amorphous accumulations of red material” [Reference Schweitzer17]. We interpret these “amorphous accumulations of red material” to be the remains of blood clots within the original dinosaur blood vessels and canals.

Other clots [36, pp. 5–7] compel us to comment on putative vascular channel blood parasites detailed recently in an isolated Brazilian sauropod dinosaur fibula. The authors describe these fusiform-bodied putative vascular parasites as “indeterminate fossilized microorganisms” and “soft-bodied parasitical microorganisms.” Histological sections show that the fibula was deeply infiltrated by calcite (and possibly sediment), as indicated by polarization microscopy [36,Figures 5A–B,D,F–H]. Researchers described this as “moderate infilling of mineral grains.” However, the calcite infiltration was widespread. The fusiform shapes, only observed within vessel canals, are reported to have a constant morphology, that is, “always showing one subtle more tapered end.” Also observed were internal “dark/more opaque spots” within the putative parasite.

It appears the Titanosaur fibula was not fixed prior to resin infiltration [Reference Aureliano36]. We routinely fix, wash, and dehydrate samples prior to embedding in hopes of preserving in detail any helminth or microbial communities within bone and to preserve the integrity of clots remaining in canals. The clots that we present here and elsewhere [Reference Armitage and Anderson44,Reference Armitage and Solliday45], in addition to clots we discovered upon UVFL examination of Camarasaurus post-cranial elements (unpublished), are generally thicker, more uniform in their close adherence to canal walls, and more responsive to UVFL microscopy, precisely because of our methodology.

The dark, opaque, and otherwise featureless nature of the fusiform shapes in the Titanosaur fibula is due to lack of fixation. Post-burial infiltration of calcite into the fibula preserved some details of the shapes, but most structural detail is lost. In contrast, a canal clot in N. lancensis (Figure 4B) features a fusiform body of a similar shape and size with clear internal structures. We interpret this fusiform microorganism in the N. lancensis vascular canal, and that shown in the horn (Figure 8), to be nematodes or other roundworms preserved by fixation. Moreover, we maintain that the Titanosaur microorganisms have been misidentified as blood parasites, which are known to be much smaller than roundworms [36, p. 10]. We have observed nematodes routinely in ground sections of our freshly-excavated-and-fixed specimens from as early as 2012, (Triceratops horridus horn, HCTH-00) [44, Figures 12,16]. Nematodes have only been observed in canals.

Interestingly, the nematode from the horn has the same “dark/more opaque spots with a larger one located centrally and a smaller one located close to the more tapered end” as described in the Titanosaur fibula [36, p. 8], but other internal details, including possible organelles, are lacking in the fibula worm. In contrast, Figure 8 shows “dark/more opaque spots” with an identical location and description as the Titanosaur finding. We maintain that preservative fixed the bodies and internal organelles of living nematodes within the Nanotyrannus and Triceratops specimens, yielding the superior preservation and detail seen here. It should not be surprising to find nematodes, given the large numbers of microbes that have been found inhabiting dinosaur bones [Reference Liang46].

The brown material immediately and closely adhering to vessel canal walls was not discussed in [Reference Aureliano36]. We interpret that brown material (more expansive in Figure 5D [Reference Aureliano36]) to be clotted blood from the Titanosaur fibula (which was probably fractured during sediment re-working, similar to the elements in this report). Significantly, we have consistently observed these clots in our dinosaur bone section preparations. They always fluoresce with UV microscopy using 347 nm illumination. Moreover, we interpret many published dinosaur bone sections to contain similar clots [36,Reference Klein and Sander49–Reference Skutschas52]. One report stated, “Vascular canals are in-filled with a dark ferric material” [47, p. 557]. It is likely that the ferric material is the remains of blood clots in the canals. Furthermore, we note that none of these theropod histological reports indicate the use of a fixative in methodology. The clots that we observe in formalin-fixed specimens are thicker, more prevalent, and more resistant to dislocation during processing than are unfixed bones.

Another researcher [53] working with fossil bones of freshwater Metoposaurs, showed multiple bone thin sections with canals filled with sediment, calcite, and pyrite. They acknowledge the presence of iron and sulfate-reducing bacteria acting on soft tissues and iron, but then they attribute infiltration of pyrite into canals to negative pressure within the bone as a result of bacterial outgassing. We wonder why original blood and blood vessels would not suffice as the source of iron and sulfates from bacterial-induced decomposition of tissues already present in vascular canals upon death.