Until recently there have been misconceptions about the existence of lymphatics in some vertebrate classes. Additionally, sparse data rules out any attempt to draw conclusions about the phylogeny of the lymphatic system; and almost all of the research on comparative anatomy was done during the 19th and early 20th century with the limitations of that period.
Already in 1919 Mayer tried to explain the contradictory findings about the piscine lymphatics by the existence of a secondary vascular system. However, only in 1981 could Vogel and Claviez experimentally prove this existence. The secondary vascular system constitutes a separate, parallel circulatory system and includes the vessels earlier assumed to be lymphatics (Hoyer 1934). It starts from the systemic arteries, forms its own capillary networks, which supply mainly the oral mucous membranes and the skin, and returns to the systemic venous system. It functions presumably in skin respiration, osmoregulation and immune defence. Based on its anatomical and functional characteristics it has been hypothesized that the secondary circulation might be an evolutionary predecessor of the lymphatic system (Steffensen and Lomholt 1992). There is evidence for a "true" lymphatic system in lungfish and it is thus reasonable to speculate that the first occurrence of a lymphatic system was associated with the transition from aquatic to terrestrial life (Laurent et al. 1978).
Growth factors and receptors that regulate lymphatic growth and development in higher vertebrates are present in fish, but their relevance for the secondary vascular system has not been analyzed (Stainier et al. 1995).
Molecules of the same growth factor family have also been identified in invertebrates, that lack endothelial cells altogether. In Drosophila these molecules direct embryonic blood cell migration. Probably only recently did these molecules assume their roles in blood vessel and lymphatic development, and it is conceivable that blood vessels evolved from blood cells (Duchek et al. 2001; Heino et al. 2001; Cho et al. 2002). In ontogenesis, however, the inverse can be observed (Ciau-Uitz et al. 2000; de Bruijn et al. 2000).
All amphibian orders are believed to have a lymphatic system. It is comparable to the mammalian system except in frogs and toads, where the superficial initial lymphatics fuse during metamorphosis to form cutaneous lymph sacs in the adult animal (Hoyer 1934; Kotani 1990). Characteristic for all amphibians are the lymph hearts, which are located at the entry points of lymph into the veins. Entry points of the lymph into the blood circulation can occur in vertebrates in three different areas: in the jugular, the lumbar and the caudal region. Most amphibian lymphatic systems communicate with the venous system in all three areas. Lymph hearts range in number from four to six in frogs to over two hundred in caecilians. Unlike their name suggests, the main function of lymph hearts is probably not the propulsion of the lymph, but rather maintaining the directionality of lymphatic flow and regulating the entry of lymph fluid into the circulation. In reptiles lymphatico-venous communications exist in the caudal and the jugular regions, but lymph hearts are found only in the caudal region (Hoyer 1934).
Mammals and most birds do not possess lymph hearts. However, unlike other vertebrates, mammals and aquatic birds possess lymph nodes. Vertebrates differ significantly in the number of lymph nodes: in humans there are between 400 and 500 lymph nodes while ducks have only four of them (Weidenreich et al. 1934). In most mammals multiple lymphatico-venous communications are formed during development. Usually only the paired communication in the jugular region persists into adulthood. Several mammalian species maintain lumbar communications into the inferior caval vein and the renal vein, draining the lymphatics of the lower extremities and the mesentery (Silvester 1912; Job 1918).
There is substantial intra-species variability in the setup of the lymphatic system. Humans have usually a paired jugular communication, but additional lymphatico-venous communications at both central and peripheral locations are not uncommon (Wolfel 1965; Threefoot and Kossover 1966; Pressman et al. 1967; Aboul-Enein et al. 1984).
It is not known whether the need for drainage was the primary selection force for the development of the lymphatics. Apart from a high-pressure cardiovascular system other evolutionary triggers might have been involved. Two of them, the hydrostatic pressure associated with the transition of aquatic to terrestrial life and the transition from poikilothermic to homeothermic temperature regulation, are associated with an increase in blood pressure, while a third one, the development of an adaptive immune system, does not require a high-pressure cardiovascular system per se.