332 M. M. Chakra varti-— Language and Literature of Orissa. [No. 4, 
of all the holy places. To supply these wants both of the pilgrims 
and of the Sevakas, various local Mahatmyas were composed in Sanskrit. 
The best known among them are— 
(a) The Viraja-Mahatmya which deals with the sacred sites of 
Jajpur. 
( b) Ekamra-Purana, Ekamra-Candrika and Svarnadri-Mahodadhi 
which describe Bhuvane 9 var. 
(c) Purusottama-Mahatmya and Niladri-Mahatmya about the Puru- 
sottama-Ksetra. 
( d ) Arka-Mahatmya about Konarka. 
( e) Kapila-Samhita, describing briefly the best known sacred sites 
in Orissa. 
The precise dates of these works cannot be ascertained. They 
themselves are silent about the times of their compositions. On the 
other hand they profess to be parts of Puranas, and to have been 
delivered by gods and Rsis in the olden time. For example, the 
Purus5ttama-Mahatmya professes to be a supplement of the Skanda- 
Purana, the Ekamra-Purana and the Ekamra-Candrika to be supple¬ 
ments of Qiva-Purana, the Viraja-Mahatmya of the Brahmanda-Purana 
and the Arka-Mahatmya of the Samba-Purana. The Puranas of 
which these works are declared to be the supplement, are all later ones, 
and have been supposed to belong to the period from the seventh to ninth 
century A.D. L The Mahatmyas must therefore be still later and cannot 
in any manner be older than the tenth century. By another line of argu¬ 
ment their time will be found to be still later. The Mahatmyas must 
be obviously later than the temples mentioned therein, and in the case 
of some temples the dates of their construction can be approximately 
arrived at. For instance the Purnsottama Mahatmya is mainly concern¬ 
ed with the temple of Jagannatha at Pari. This great temple, as I have 
already proved 1 2 was built under the orders of Codagagga, the founder 
of the Grai)gavam 9 a dynasty. Its date thus comes to about 1140 A.D. 
One hundred years may be fairly taken as the period during which 
the human origin of the temple was forgotten, and a divine origin with 
1 These Puranas do not fulfil the five lalcsanas required by the Amarak 59 a and 
must therefore be later than the sixth century A.D. 
[The supposition that the Author of the Amarak 59 <a lived in the 6 th Century 
A.D., rests merely on the well-known tradition of the nine Jewels at the conrt of 
Vikramaditya. On the date of the AmarakStja see now : Zachariae, Die Indishen 
Worterbucher (ko<ja) in Bidders Encyclopaedia of Indo-Aryan Research, I, 3. B. 
page 18, Ed.] 
2 Journal Asiatic Society of Bengal, Yol. LXIV., Pt. I, No. 2, pp. 130-1, 
135. 
