"He used to call the female horse a mare." It was recorded by al-Ḥākim (2/144) through the route of ʿUthmān ibn Saʿīd al-Dārimī, who reported from Mūsá ibn Sahl, who reported from Marwān ibn Muʿāwiyah al-Fazārī, from Abū Ḥayyān al-Taymī, from Abū Zurʿah, from Abū Hurayrah as a marfūʿ ḥadīth, and he said: "Authentic according to the criteria of al-Bukhārī and Muslim," and al-Dhahabī agreed with him! I say: No, for this Mūsá ibn Sahl was not recorded by al-Bukhārī and Muslim at all, and among the narrators in his class, there are four, all named Mūsá ibn Sahl. The first: Mūsá ibn Sahl ibn Qādim al-Ramlī, originally from al-Nasāʾī, and he is trustworthy, from the shaykhs of Abū Dāwūd and al-Nasāʾī. The second: Mūsá ibn Sahl ibn Kathīr al-Washshāʾ al-Baghdādī, and he is weak. The third: Mūsá ibn Sahl ibn Hārūn al-Rāzī, very weak. The fourth: Mūsá ibn Sahl al-Rāzī, unknown. It is not clear to me now which of them is the narrator of this ḥadīth, and until it becomes evident that he is trustworthy, it remains weak, and Allāh knows best. Then I found a corroborating narrator for him, as Abū Dāwūd said (1/399): Mūsá ibn Hārūn al-Raqī reported to us, Marwān ibn Muʿāwiyah reported it. Al-Raqī, from whom Abū Ḥātim al-Rāzī and other imams heard, and Ibn Ḥibbān mentioned him in "al-Thiqāt". Thus, with this corroboration, the ḥadīth is saved from weakness and enters the category of authentic or at least ḥasan. And Allāh knows best.