XXXVl 



PREFACE. 



Letters. 



Genders, 



Gorman way of talking, the second t is " inorganic," 

 and serves only to mark the shortness of the vowel. 

 Under this form the word is our Twit. 



Enough has been said to show that the length of 

 the vowels in Saxon English is a very wide subject, 

 and to justify the postponement of any decisions in 

 the Glossary. 



In our oldest manuscripts jwjin often occurs where 

 it is the custom to print T. Rejp'S, led, rest, Luj-^, 

 'pleasure, lust, and a hundred others are examples : 

 the superlatives end in J^orn, as ^ ag^eley^e mseben, 

 the very noble maiden, the participles also. In the 

 Codex Exoniensis the editor removed these features 

 of antiquity ; they offended him ; and wore not ac- 

 cording to Rask.^ If any such occur in the present 

 volume they are preserved ; they are not dialectic, 

 but archaic. 



In genders the glossaries are untrustwortliy ; thus, 

 the most recent is found, as regards the few words 

 common to both, much wrong, when compared with 

 the citations in that at the end of this volume. It 

 is unsafe to trust compounds with je-, for the gen- 

 ders of the simples, for Ge- being a form of Con- and 

 collective, its com])ounds are found to have a tendency 

 to run into the neuter." Simples cannot always be relied 

 on for the gender of the compound ; all moderns take 

 poppyjib for a feminine, after pyj^^, but in a wide scope 

 of unpublished materials I have always found it neuter.^ 

 Occasionally a new principle comes in, and by attrac- 

 tion the article agrees with the former element in the 

 compound, instead of the latter ; hence pserejieebpe 



' For example, Gebiej-jaS, Gepel- 1 •' Tja cncoj>holen, Lb. I. xlvii. .3, 



Sa'S, p. .358; I'cob', p. 3.57. Ahpe- 

 ol'eS, p. :r.)7 ; Blse'S, p. .",10. 



- Thus Sp)i£ec is feminine, fJe- 

 I'ppxc, neuter. 



perliaps makes kneeholly neuter ; 

 or else Tpa, is tivo parts. This 

 remark slioukl have appeared in the 

 Glossary. 



