Historically Edwardsiella tarda has been a bacterial pathogen of nominal concern in farmed catfish in the USA. In the 2010s, E. tarda was reaffiliated into three species: E. tarda, E. piscicida, and E. anguillarum, of which E. piscicida is recognized as a significant, emergent pathogen within the industry. This reorganization has obscured previously documented disease pathologies of these bacteria in catfish. This study clarifies discrepancies in pathologic changes and virulence of E. tarda, E, piscicida, and E. anguillarum in channel (Ictalurus punctatus), blue (I. furcatus), and channel x blue hybrid catfish and assesses potential cross-protection of Edwardsiella congeners against subsequent E. ictaluri and E. piscicida challenge. Channel, blue, and hybrid catfish (20 fish/tank; 3 tanks/treatment) were intraperitoneally challenged with two (E. piscicida; low, high) or three (E. tarda and E. anguillarum; low, medium, high) doses of E. tarda, E. piscicida, and E. anguillarum. In light of preliminary data suggesting increased virulence in catfish, the highest dose of E. piscicida was roughly equivalent to the lowest doses of E. tarda and E. anguillarum. Bacterial and histopathology samples were collected at 1, 2, 3, 5, 10, and 14-days post challenge (dpc). Cumulative mortality was recorded for 14-dpc in separate non-sampled tanks (20 fish/tank). Cumulative mortalities ranged from 25-95%, 20-55%, and 80-100% in hybrids, channels, and blues, respectively, challenged with E. piscicida. Conversely, E. anguillarum and E. tarda produced ≤5% mortality, regardless of catfish species, at doses equivalent to the highest dose of E. piscicida. Gross changes included poor body condition, mild ascites, transmural body wall ulcerations, and, rarely, dorsocranial ulceration (hole-in-the-head). Histologic lesions were characteristic of Gram-negative sepsis, with more severe lesions in fish infected with E. piscicida, regardless of dose, and the highest doses of E. anguillarum and E. tarda. Predominant early lesions included disseminated hemorrhage and necrosis often with numerous bacteria accompanied by gastric and intestinal submucosal edema. Transition to a granulomatous response with reduced bacterial numbers began around 3-dpc and was the dominant lesion by 5-dpc. Lesions were uncommon in all fish surviving 14-dpc. Exocrine pancreatic degeneration occurred sporadically, with most fish affected between 3-dpc and 10-dpc. Coelomitis occurred frequently, likely secondary to the intraperitoneal infection route. Fish infected with medium doses of E. tarda and E. anguillarum occasionally had similar but milder tissue changes. Channel, blue, and hybrid catfish (n = 9-25 fish in a single tank) that survived previous challenge with E. piscicida, E. anguillarum, or E. tarda were exposed 100 days after the initial challenge to E. piscicida via intraperitoneal injection or E. ictaluri via 30-minute immersion baths. Hybrids previously challenged with E. piscicida or E. anguillarum were significantly more likely to survive E. piscicida infection than naïve controls (p < 0.05). Similarly, channels were significantly more likely to survive an E. ictaluri infection (p < 0.05). Fish that survived E. tarda challenge were not protected against subsequent challenge with E. piscicida or E. ictaluri (p < 0.05). These results support observations that E. piscicida is more pathogenic in catfish compared to E. anguillarum and E. tarda. Cumulative mortality varied between catfish species, with blue and hybrid catfish more severely affected, supporting previous work indicating increased susceptibility of hybrid and blue catfish to these agents. Lastly, this work supports research evincing a cross-protective effect among some Edwardsiella congeners. Further work is warranted to elucidate the cross-protective nature of the immune response against E. anguillarum and E. piscicida to subsequent E. ictaluri and E. piscicida challenge.