This is in harmony with the following statement of R. Abdimi b. Hama b. Dosa: What is the significance of the text: It is not in heaven, [that thou shouldst say: ‘who shall go up for us to heaven, and bring it unto us’], neither is it beyond the sea [that thou shouldst sat, ‘Who shall go over the sea for us, and bring it unto us’]? ‘It is not in heaven’, for if it were in heaven you should have gone up after it; and if it were ‘beyond the sea’, you should have gone over the sea after it. Raba expounded, ‘It is not in heaven’, it is not to be found with him who, because he possesses some knowledge of it, towers in his pride as high as the heavens, ‘[neither is it beyond the sea’] it is not found with him who, because of some knowledge of it, is as expansive in his self-esteem as the sea. R. Johanan expounded: ‘It is not in heaven’, it is not to be found among the arrogant; ‘neither is it beyond the sea’, it is not to be found among merchants or dealers. Our Rabbis taught: How are the sabbath boundaries of towns extended? [If a town is] long the sabbath limits are measured from its normal boundaries. If it is round corners are added to it. If it is square no corners are added to it. If it was wide on one side and narrow on the other it is regarded as if both its sides were equal. If one house projected like a turret, or if two houses projected like two turrets, they are to be treated as if a thread had been drawn beside them in a straight line, and the two thousand cubits are measured from that line outwards. If the town was shaped like a bow or like a gamma, it is to be regarded as if it had been full of houses and courtyards, and the two thousand cubits are measured from the imaginary boundaries outwards. The Master said: ‘[If a town is] long the Sabbath limits are measured from its normal boundaries’. But is this not obvious? — The ruling is required in a case where it was long but narrow. Since it might have been presumed that the width should be regarded as equal to its length. we were informed [that the law was not so]. ‘If it is square shaped no corners are added to it’. Is not this obvious? — This was only required in a case where it is square shaped but the sides of the square are not parallel with the four directions of the world. As it might have been presumed that it should be deemed to be enclosed in an imaginary square whose sides are parallel with the four directions of the world, we were informed [that this is not permitted]. ‘If one house projected like a turret, or if two houses projected like two turrets’. Now that you said that the law applied to one house, was it also necessary to mention two houses? — The ruling was necessary in that case only where the two houses were respectively on two sides of the town. As it might have been presumed that we apply the law only where a projecting house was on one side but not when houses were projecting on two sides, we were informed [that the law is applied to the latter case also]. ‘If the town was shaped like a bow or like a gamma, it is to be regarded as if it had been full of houses and courtyards, and the two thousand cubits are measured from its imaginary boundaries’. R. Huna laid down: If a town is shaped like a bow, then, if the distance between its two ends is less than four thousand cubits, the Sabbath limits are measured from the bow-string, otherwise measuring must begin from the arch. But could R. Huna have laid down such a ruling? Did not R. Huna in fact rule: If a breach was made in a town wall, [the houses on both sides of the breach are regarded as belonging to the same town if the distance between them is] no more than a hundred and forty-one and a third cubits? — Rabbah b. ‘Ulla replied: This is no difficulty, since the former deals with a case where the gap was only on one side while the latter deals with one that had breaches on two sides. Then what does he inform us? That a karpaf is allowed for each section. But did not R. Huna once lay down such a ruling, as we learned:ᵃᵇᶜᵈᵉᶠᵍʰⁱʲᵏˡᵐⁿᵒᵖᵠʳˢᵗᵘᵛʷˣʸᶻᵃᵃᵃᵇᵃᶜᵃᵈᵃᵉᵃᶠᵃᵍᵃʰᵃⁱᵃʲᵃᵏᵃˡ